Grey wisdom? Philosophical reflections on conformity and opposition between generations
edited by roel van goor & ernst...
15 downloads
690 Views
1MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Grey wisdom? Philosophical reflections on conformity and opposition between generations
edited by roel van goor & ernst mulder
Grey wisdom? Philosophical reflections on conformity and opposition between generations
GREY WISDOM? PHILOSOPHICAL REFLECTIONS ON CONFORMITY AND OPPOSITION BETWEEN GENERATIONS
Edited by Roel van Goor &
Ernst Mulder
The publication of this book was supported by the Department of Education, Faculty of Social and Behavioural Sciences, of the University of Amsterdam
Lay-out: Ernst Mulder, Naarden Cover design: René Staelenberg, Amsterdam ISBN 90 8555 351 2 NUR 849 © Amsterdam University Press, 2006 All rights reserved. Without limiting the rights under copyright reserved above, no part of this book may be reproduced, stored in or introduced into a retrieval system, or transmitted, in any form or by any means (electronic, mechanical, photocopying, recording or otherwise) without the written permission of both the copyright owner and the author of the book. Alle rechten voorbehouden. Niets uit deze uitgave mag worden verveelvoudigd, opgeslagen in een geautomatiseerd gegevensbestand, of openbaar gemaakt, in enige vorm of op enige wijze, hetzij elektronisch, mechanisch, door fotokopieën, opnamen of enige andere manier, zonder voorafgaande schriftelijke toestemming van de uitgever. Voor zover het maken van kopieën uit deze uitgave is toegestaan op grond van artikel 16B Auteurswet 1912 jº het Besluit van 20 juni 1974, Stb. 351, zoals gewijzigd bij het Besluit van 23 augustus 1985, Stb. 471 en artikel 17 Auteurswet 1912, dient men de daarvoor wettelijk verschuldigde vergoedingen te voldoen aan de Stichting Reprorecht (Postbus 3051, 2130 KB Hoofddorp). Voor het overnemen van gedeelte(n) uit deze uitgave in bloemlezingen, readers en andere compilatiewerken (artikel 16 Auteurswet 1912) dient men zich tot de uitgever te wenden.
Contents
Preface Roel van Goor & Ernst Mulder
7
1. On the value of thinking about goods and values in education Doret de Ruyter
9
2. Searching for the good concerning education. Response to Doret de Ruyter Christiane Thompson
25
3. The changing structure of scholarly communication in educational science Raf Vanderstraeten
31
4. The discontinuous development of scientific disciplines: Some remarks on scholarly authorship. Response to Raf Vanderstraeten Roel van Goor
47
5. On the shoulders of giants Christopher Winch
53
6. On the grounds of arguments. Response to Christopher Winch Leonie le Sage
75
7. Education, history, and human nature Jan Bransen
81
8. Education from the inside and the outside. Response to Jan Bransen Judith Suissa
99
9. The student Klaus Mann. Betwixt and between individualism and ‘generation-unit’ Brita Rang
105
10. The constitution of the subject and the possibility of critique. Response to Brita Rang Stefan Ramaekers
125
11. The contingency of scholarly development across generations Frieda Heyting
131
List of contributors
139
Preface ROEL VAN GOOR ERNST MULDER
This book contains the texts that were discussed during a conference with the same title. The contributions to the conference were written, edited, and published in advance, and served as a starting point for deliberation. At first sight this book appears to be a standard volume published on the occasion of a meeting of scholars, gathered round a theme of scientific interest. A closer look at the chapters reveals that the authors are all philosophers of education, and that every two articles make a couple. Every pair consists of a statement of a sometimes elaborate argument, concerning the central theme of the book, accompanied by a shorter, but nevertheless thorough, critical response. Furthermore, one scholar seems to be referred to more than any other one, and she also turns out be the author of the last piece in the book. That is no coincidence: Frieda Heyting, or rather her retirement as a professor in the philosophy of education, is the immediate cause of the conference and of the book. But make no mistake: this is not a liber amicorum in the usual sense of the word. This book is no attempt to laurel the professor emeritus-to-be, it contains no overview of all her contributions to the field of the philosophy of education, and also no list of her publications as a scholar. As far as her work is mentioned, it is usually in the context of the other authors disagreeing with her. They are friendly, but very critical in dealing with her scientific achievements. Which is, incidentally, very much in line with her own position as a ‘doubting’ sceptic.1 The main characteristic of the conference, and of the book, is a critical consideration of arguments, as should always be the case in the philosophy of education, as well as in any other academic area. The title of this book, ‘Grey wisdom? Philosophical reflections on conformity and opposition between generations’, deserves some elucidation. 1
See her own contribution in this collection: ‘The contingency of scholarly development across generations’.
7
Generation’ is a muddy concept2, so why would anybody like to use it as a banner for a conference? First, muddy concepts demand clarification, and this is one of the responsibilities of a philosophical approach. And ‘generation’ may be a slippery concept, but it is also most definitely educational. Then there is the issue of the succession of generations, and the ways in which one generation, usually the older, influences the others. Should younger generations accept the inherited achievements and traditions, or should they rather focus on developing something new? Is there any escape form the great chain of generations, or do the young inevitably act like their predecessors? Does the answer to this question apply to academic generations as well as to generations of children and adults, or is there maybe a difference in this respect between science and education? This book offers, among many other things, an empirical touchstone to questions like these, because the relationship between generations is built into its design: the critical comments are all written by members of the younger academic generation. This is, we think, what Frieda Heyting has in mind when she describes herself as a successor to her teachers, and leaves it up to the younger generations to cope with the traditions they inherit. As she states in her contribution to this book: “I am sure they will decide for themselves what to do with it.”3
2
See Rang in this volume: ‘The student Klaus Mann. Betwixt and between individualism and “generation-unit”’. 3 Heyting in this volume, p. 137..
8
1 On the value of thinking about goods and values in education DORET DE RUYTER
Introduction When I received the invitation to write a contribution to a book on the theme ‘Grey wisdom - conformity and/or opposition between generations’ and the suggestion that authors reflect on questions regarding educational practices as well as their own position as academics in relation to former or consecutive generations, my first reaction was that I was uncertain what to make of the second question. It seems obvious that in the development of an academic discipline there is always both the quest for originality of a new generation and a building on the wisdom of the former generation. Is it not selfevident therefore that good academics acknowledge that they are indebted to their predecessors? The first question, however, i.e. the question what one generation should or may pass onto the next, I believe to be one of the key questions of philosophy of education. Wise answers have already been given, but, and this mirrors the argument I just gave, every age has her own, new, questions for which given answers are not sufficient. Philosophy of education will therefore, as I will stress again at the end of this contribution, always be of vital importance to the academic as well as practical disciplines within education. I consider myself a representative of the generation of philosophers of education that follows Heyting’s generation and I am therefore pleased that at least one of the academics of the ‘in-between generation’ who contribute to this work has been educated within philosophy of education in the Netherlands. That I am the only one is not due to the fact that there were not sufficient PhD students in the past decade and a half. Many PhD students have graduated in philosophy and history of education and quite a number of them have written impressive theses. Nevertheless, hardly any one of them has been able to find a job at a university and philosophy of education is not par9
ticularly thriving within Dutch universities at the moment. Why this is the case, is not something I will deal with in this contribution, but I will try to show why this work is necessary for the academic discipline of education. Despite my qualms about the second question, this contribution will address both afore mentioned questions. First, I give a concise sketch of the history of philosophy of education at the Vrije Universiteit and describe in which paradigm I position myself. The remainder, which is also the main and principal part of this contribution, is an example of the way in which I work and write as a philosopher of education and consists of an analysis and justification of the normative question what the current generation of educators should offer to children. I argue that good education consists in offering goods and values to children and that good educators acknowledge that these can have a variety of conceptions or interpretations. Therefore, they need to assist children to become adults who are able to reflect upon the conceptions possible and gradually give children freedom to explore which precise content they wish to give to the goods and values.
My predecessors With regard to the position of values in particular, there has been a radical change in philosophy of education at the Vrije Universiteit. The first professor in education, who was also responsible for Philosophy of Education, Jan Waterink, argued that the academic study of education is always normative (1958, p. 30). In his view, every theory of education is based on a priori premises; they only differ in the nature of these premises. For instance, where Waterink’s own a priori was explicitly Calvinist, he believed that those who based their theory on scientific insights equally professed a statement of faith, theirs was simply different in content (idem, p. 48). With the appointment of Ben Spiecker as professor in philosophy of education in the middle of the seventies and Jan Steutel as associate professor at the beginning of the eighties, the wind started to blow from the West. Their approach implied a shift in paradigm at the Vrije Universiteit. They positioned themselves within the Anglo-Saxon tradition, in which research consists primarily of conceptual analysis as well as the analysis of premises underlying educational practices and theories and normative claims about those practices and theories. 10
The analytic-philosophical approach has been accused of concentrating too heavily on the analysis of concepts. I underwrite the claim that clarification of concepts is important, but not sufficient for addressing ethical issues within philosophy and philosophy of education (e.g. Murdoch, 1970). Philosophy of education also needs to make normative claims. To be fair, most analytically oriented philosophers of education have done so in the past. One of the most esteemed representatives, Richard Peters, did not shy away from normative claims about the content of the aim of education as well as educational practices. His theory about the educated person, for instance, is not only a normative claim about the aim of education, but also implies normative criteria with regard to the educational practice that contributes to the development of children into educated persons (see for instance, 1981). That he justified these with conceptual as well as internal pedagogical arguments does not make them descriptive. Spiecker and Steutel have also made normative-ethical claims, particular in their latest works, for instance with regard to the question whether or not people with learning difficulties should be allowed to have sex (2002), or with regard to the topic of civic education and its implications for the identity of people from collectivistic cultures (2004). Although it is possible to conceive of philosophy of education or education in general as a descriptive academic discipline, I would want to defend the importance of normative-ethical reflection and justification, precisely because the practices upon which it reflects are value laden and because these practices can be judged to be good or not. The fact that this is possible gives academics, since they are in a privileged position due to their knowledge and insights, the responsibility to contribute to the evaluation of goodness of practices by generating knowledge as well as normative ethical reflection upon that knowledge. Therefore, while conceptual analysis and empirical research are necessary, for conceptual analysis clarifies the words we use in educational theory and practice and empirical research describes and interprets reality, they are not sufficient. Academics also need to address normative-ethical questions, for instance via the method of the reflexive equilibrium, which aims for an optimal coherent balance between one’s well-considered judgements and intuitions, ethical principles as well as philosophical insights and empirical data from diverse disciplines (Van der Burg en Willigenburg, 1998; Spiecker en Steutel, 2001). The requirements of the method of reflexive equilibrium, which are relevant for all criticalsystematic normative-ethical research, namely that the arguments be public in character, that it comprise critical reflection about the premises and that 11
the premises can be changed in confrontation with other insights, are not necessarily characteristic for a normative philosophy of education, and certainly not for my pre-predecessor, Jan Waterink, who perceived it as a theory construction based on unchangeable foundations. I follow the tradition of my academic masters Spiecker and Steutel and position myself within the analytic tradition and will, as said in the introduction, exemplify this by addressing the main question of this contribution: What is good education and what makes it so?
Good education The way in which adults relate to children and assist them in their development towards adulthood has historical continuity as well as change. The continuity concerns the wish of parents that their children will do well in life, which is expressed in their efforts to gratify the basic needs of children, like food, shelter and care as well as their need for schooling. The change lies in the ways in which adults attempt to meet the needs of children as well as the perception of children that underlies their actions (see for instance the special issue of Justitiële Verkenningen, 2005). Education is a teleological activity; it comprises actions of adults that aim to assist children to develop into adult persons. It is therefore possible to say that good education means ‘good’ for its purpose or goal, i.e. good education is instrumentally good (Von Wright, 1963). The conceptions of good education that reside under this concept can be distinguished on the basis of both the aim as well as the means or process. Although there is not a necessary relation between the two, i.e. the same aim can be pursued by diverse means and the same means can be used to pursue different aims, the relation is not merely contingent. It is obvious that not all means and ends can be combined. If educators aim for loyal docility or what Callan (1997) calls ethical servility, they will not stimulate children to think critically, or alternatively, if educators have the aspiration that their children become active citizens in a liberal democracy they cannot indoctrinate their children into a particular conception of the good life. This concept of good education is without further specification or content a tautology. It only states that education is good if it serves its purpose well. The interesting question is obviously, what is its purpose? In my view, education can be called good if it contributes to the possibility of children to de12
velop into flourishing adults. Where minimally acceptable education, i.e. the level that is seen as the threshold level below which state intervention is justified, is conducive to the development of children into self-providing and non-dependent adults, by fulfilling the basic needs of children, good education aims for something more or higher. But what does human flourishing consist of and what are the implications for education? This question seems to have become more complicated and maybe also more urgent in current western societies. The fall of the grand narratives has resulted in a fragmentation of value systems in Western societies or what Baumeister and Muraven call ‘the value gap’: “a cultural deficit which presents individuals with a problem in how to find meaningful ways to endow their lives with value and make choices that are right and good” (1996, 409). Are there goods of which we can be certain they will contribute to the flourishing of children?
Goods and values The idea that there are goods that are objectively good for all is controversial, which is related to the contention that there are characteristics of human beings that are essential to them (only). Although a lot can and has been said about this, I propose to accept, what seems to me the most reasonable position, that there are goods that are related to human characteristics and therefore need fulfilment in order to contribute to their well-being. This does not imply that they are good for humans only (some can also be good for animals or for non-human entities) and therefore would be the characteristics that are distinctive for the human species. Neither does the fact that the goods are good for all human beings, because they all share the same characteristics, mean that they are good for all in the same way. There are basic human needs, related to biological characteristics of human beings, that need fulfilment in order for people to be able to flourish. Meeting these needs could be said to be part of minimal education. These would be health, safety, freedom, sufficient means for existence, the possibility to develop (physiologically as well as psychologically). If these needs are not met, we can be certain that persons will have grave difficulties leading a flourishing life. Good education consists not only in fulfilling these needs whilst children are young, but also assists children learning to reflect on which way(s) of fulfilling these needs would be good for them when they 13
are adults. For, there are various ways in which one can lead a healthy life, can develop or earn a sufficient income. I will return to this issue several times in this contribution. Additionally, there are basic psychological human needs whose fulfilment is necessary for human flourishing. In psychological theories of human flourishing, a distinction is being made between hedonic and eudaimonic theories of well-being, which is similar to the distinction that philosophers make between subjective and objective accounts of human well-being. The eudaimonic tradition describes human well-being as personal expressiveness and self-realization “where what is considered worth desiring and having in life is the best within us or personal excellence” (Waterman, 1993, p. 679). The Self-Determination Theory of Ryan and Deci is currently one of the most influential theories within the eudaimonic tradition. This theory aims to clarify what it means to actualise or realise the self and how that can be accomplished (Ryan & Deci, 2001, p. 146). Ryan and Deci argue that there are three innate psychological needs or nutriments that are functionally essential to a person’s flourishing (Reis et al., 2000, p. 420). These needs are the need for autonomy or self-determination, for competence or effectance, and for relatedness or affiliation. These needs have been found to be universal, although this does not imply that “their salience and their avenues for satisfaction are unchanging across the life span or that their modes of expression are the same in all cultures” (Ryan and Deci, 2000, p. 75; see also Ryan and Deci, 2001, p. 147). Where the need for competence is fulfilled “by the experience that one can effectively bring about the desired effects and outcomes1” (Reis et al., 2000, p. 420), autonomy is described as “an inner endorsement of one’s actions, the sense that they emanate from oneself and are one’s own” (Deci and Ryan, 1987, p. 1025). Defined this way, autonomy does not conflict with relatedness, the second basic need, which “pertains to feeling that one is close and connected to significant others” (Reis et al, 2000, p. 420). Autonomy is not the opposite of dependence, but of heteronomy “in which one’s actions are experienced as controlled by forces that are phenomenally alien to the self or that compel one to behave in specific ways regardless of one’s values or interests” (Chirkov et al., 2003, p. 98). As Chirkov et al. (2003) explain, the SDT assumes that there are two axes that are orthogonal to each other: autonomy versus heteronomy and independence versus dependence. 1
This is also defended by Baumeister (1991) who on the basis of a wide range of examples, establishes that when people lose control or influence they lose meaning in life and will develop feelings of depression.
14
The genesis of one’s action can reside in one’s loyalty to one’s parents, partner, children or friends. For instance, the fact that parents act in ways that serve the interests of their children more than their own does not mean that they are not autonomous. If they feel that they are the owner of their actions, they may willingly give up their independence and take decisions that may even be detrimental to their own interests. The empirical researches Ryan and Deci conducted with others also indicate that it is more important to focus on these needs than on what persons might want. Ryan has found with Kasser, for instance, that the vigorous pursuit and feelings of efficacy with respect to achieving wealth and fame are actually associated with ill being, more than with well being (Kasser & Ryan, 1993). This has implications for a seemingly predominant way in which people currently try to find meaning or security: within (in) the economic domain. Ritzer, for instance, argues that one of the reasons for the success of a chain like McDonald’s is the fact that it is predictable. This may be an advantage: “in a rapidly changing, unfamiliar, and seemingly hostile world, the comparatively stable, familiar, and safe environment of a McDonaldized system offers comfort” (Ritzer, 2000, p. 16). Similarly, we can observe that people try to construct their identity on the basis of consumer products. Product developers have become masters in convincing people that by buying a particular kind of trainers or car, one gets a supplementary identity. Advertisements do no longer address the purpose and characteristics of a product, but picture the lifestyle or identity that the product signals. Although I do not want to claim that educators should try to eliminate the pursuit of extrinsic and particularly materialistic aims in their children, for they do need material goods in order to be able to flourish, I would wish to defend not only that children should learn that material goods in themselves will not bring happiness (which seems so obvious that it is trivial) but also that they should only pursue these in ways that do not endanger the fulfilment of other basic interests, for instance their health, or their need for relatedness. Given the empirical evidence of the claim that the mentioned psychological needs contribute to human flourishing, which was found and confirmed in various reliable and valid researches of which I quoted several above, we may argue that good education consists of assisting children to develop ways in which they can fulfil these needs. There are two comments to be made to this position, which are both related to the general character of the needs. First, it should be stressed that educators can only be held responsible for their efforts to provide the conditions that make it possible for children to 15
flourish. They cannot ensure that children will flourish as adults. Not only is it impossible for them to foresee the future or control the lives of their children as adults, but flourishing depends on persons themselves: only if persons are able to fulfil the needs in a way that is good for them, will they flourish. Secondly, although educators can be certain that the fulfilment of these needs is good for children as youngsters and adults and therefore should feel obliged to foster the development of children’s capacities to fulfil these, they cannot prescribe how children will come to fulfil them, for the ways in which these needs can be fulfilled vary, and children need to discover for themselves which way might best contribute to their flourishing. This brings me to the question of values in education. For, in order to give an interpretation that is good (prudentially as well as morally), persons need values. I follow Frankfurt who has, in my view, convincingly argued that persons need a relatively stable framework of values or ultimate ends in order to make one’s experiences and one’s life intelligible and valuable (Frankfurt, 1999, p. 85). “Unless a person makes choices within restrictions from which he cannot escape by merely choosing to do so, the notion of selfdirection, of autonomy, cannot find a grip” (Frankfurt, 1999, p. 110). This leads to the following, possibly even more contentious, question, namely which values may educators offer to children? An answer to this question that I have become to find sympathetic, is Joseph Raz’s theory of values. Joseph Raz defends a social dependence thesis, which consists of two theses: “The special social dependence thesis claims that some values exist only if there are (or were) social practices sustaining them. The (general) social dependence thesis claims that, with some exceptions, all values depend on social practices either by being subject to the special thesis or through their dependence on values that are subject to the special thesis” (2003, p.19). According to Raz, values exist where there is a practice that sustains those values, even though people may not be aware that their acting is based on the values that underlie the practice. The way in which Raz perceives of the relation between value and practice is on the one hand in terms of a necessary relation, i.e. a value will only come into existence within a social practice, but on the other hand seemingly contingent as he claims that values are not dependent on the original practice to remain of value; practices are necessary for the origin of values, but not for their continuity. Practices can change without the values necessarily disappearing. This means that Raz’s theory is not a form of social relativism, because the values are only dependent on social practices for coming into existence, not for their recognition afterwards or by others. Neither is his theory conventional, 16
he claims, because a practice is not the reason for the value of the value, it is the reason for its existence. Although there may be disagreement about the ways in which standards or concepts are interpreted, such disagreements affirm the objectivity of values, according to Raz, for “the disagreements are contained within a framework of shared views” (2003, p. 51). The special dependence thesis seems to apply primarily to cultural values, “because sustaining practices are a necessary condition for it to be possible for these values to be instantiated, and the possibility of instantiation is a condition for the existence of values” (Raz, 2003, p. 33, 34). There are four classes of values whose instantiation do not depend on a sustaining practice, namely a) pure sensual and perceptual pleasure; b) aesthetic values of natural phenomena; c) many enabling and facilitating values like freedom or justice, and d) the value of people, and of other ‘valuers’ who are valuable in themselves. The last two categories, however, can be related to the special dependence thesis via the general dependence thesis: they only have a point or are of value if they enable people to value the socially dependent values. The value of these values is therefore, at least partly, dependent on the social dependent thesis2. The social dependence thesis leads to value pluralism i.e. that there are many distinct values and there are incompatible values, but not to relativism. The ground for saying that an action or object is good is relative to a particular genre, but this verdict is unrestricted, i.e. it is absolute. A piece of art, but also social arrangements can be judged with criteria that belong to their genre. Raz offers the example that one judicial system is good because it is a good adversarial system, whereas another can be a good prosecutional system. Both are good absolutely but good in different ways. This allows us to say without contradiction that works of art or social arrangements that are completely different are both good. Children should be taught the standards of genres, not only in order to become good judges of values, but also to enable them to give a valuable interpretation of the goods that are conducive to their well-being. As I have already mentioned, and will deal with in the next section, the general goods 2
With respect to the moral values that reside under c and d, particularly those under d, it seems intuitively wrong to give these values an instrumental status. Are they not of intrinsic value? Raz argues that we do believe that they are a-temporal or eternal, but “they have a point only under certain circumstances. For most values their point depends on it being possible to recognize them and engage with them. They are idle and serve no purpose if this is impossible. In this sense the value of valuers depends on other values, for what is special about valuers qua valuers is their ability to engage with values” (Raz, 2003, p. 36).
17
need interpretation. In Raz’s terms: the goods have different genres that can all be good. Which genre will be good for children is something they have to discover themselves, but what makes their actions within this genre good is something that educators can teach them. As John White recently argued, “Children’s education in this area should induct them into the goods themselves; but it should also lead them into a reflective appreciation of the nature of these goods, their provenance, historical development and variety. Only in this way can they, too, become members of the class of reliable judges about the nature of personal well-being and its values” (Raz, 2003, p. 268).
Good education Good education, then, first of all consists in offering children the goods that will be conducive to their well-being. Secondly, because these goods are, as said, general, educators need to assist children in developing the capacities to think about the way in which the goods will be good for them and at least not detrimental for others. The first educational task of educators, i.e. ensuring that children know which goods they need to pursue to be able to flourish, does not seem to be a reason for controversy, because the goods are good for all, they are formal and general. However, precisely because they are formal and general, educators cannot suffice with ‘simply telling which they are’. Educators need to concretise these goods by reference to examples, by reacting to children’s behaviour and that of others, etc. Another way in which educators pass on the goods to children is by exemplifying the way in which the goods contribute to their flourishing. Good education may therefore be endangered if parents are unable to do so, because of external circumstances or internal obstructions like severe depression or psychosis, or unwillingness. For instance, not all parents are willing to give up smoking if they have children, even though they know that this is detrimental for their children now and not serving them as youngsters either, because it has been proven that children of parents who smoke will have a higher chance of becoming smokers themselves (and parents cannot pretend ignorance after all the public campaigns). The second educational task is related to the aim of assisting children to become practically wise adults who are able to give meaning to the goods in a way that is good for them and not detrimental to others. The goods are 18
complex for two reasons: they are general and require adaptation in order to have significance in a person’s life; they are what Raz calls mixed values, i.e. “values, constituted by standards determining ways for ideal combinations of contributing values, and criteria for various relationships that objects can have to them” (2003, p. 49). For instance, the way in which one gives meaning to the good of being related to meaningful others can be highly diverse. Where one person flourishes by having a spiritual relationship with a deity, others thrive by having a long-term commitment to a person of the same or different gender and again others prosper by not having an exclusive relationship at all. Being wise assumes more than being intelligent or knowledgeable. A practically wise person can “deliberate finely about what is good and beneficial for himself, not about some restricted area - e.g. about what promotes health or strength - but what promotes living well in general” (Aristotle, 1140a, 25-30). A practically wise person possesses understanding (Raz, 2003). Understanding in Raz’s view means that a person is able to place an object in its context and her knowledge of what is understood is connected to her imagination, emotions, feelings and intentions (2003, p. 48). In other words, a practically wise person is not only a knowledgeable person, but also someone who is attached to what she understands to be good, right or true and has the will to act upon her insights. For instance, with regard to the good of health, a practically wise person will be a modest or frugal person. Although she does not have to eat healthily all the time and might give in to the temptation to eat fries and a Whopper now and then, she will think about her diet carefully - not only with regard to herself but also in relation to ecological and socio-economic issues (how is food produced, by whom is it produced). Developing into a practically wise person takes a long time and demands diverse qualities of educators, for instance that they are a good example and therefore themselves practically wise adults and that they are dedicated educators who assist children to become virtuous and wise by frequently offering children the opportunity to practice their developing virtues and by responding to their actions systematically, i.e. who habituate children (e.g. Steutel & Spiecker, 2004). Finally, educators should give children freedom to explore which of the interpretations or genres is good for them. If parents aim to ensure that their children adopt their way of life or conception of the good that has shown to fulfil their own basic needs, they do not enhance their children’s ability to satisfy the need of autonomy and competence. Their children may do as their 19
parents wish, but by doing so they act out of extrinsic reasons and are not internally motivated, which will reduce their feeling of well-being (see for instance, Deci, et al 1996). Good education is characterised by a growing amount of freedom in order to meet their mentioned need for autonomy. For young children unlimited freedom would be too dangerous and therefore a risk for their development into flourishing adults. Additionally, habituation limits children’s freedom. But, while children are developing these abilities, educators need to give children the possibility to explore. If children are not given this opportunity, they will not only be hindered to become practically wise, because they do not act upon their own insights but on their educators’, but it is also likely that they will not flourish, because they live a life that is good for others but not necessarily good for them. Thus, good education can be characterised as an education in which parents and professional educators pass on goods and values and teach children the capacities to reflect upon these to discover which conception of the good would contribute to their own flourishing. Would education be even better if educators stimulate children to think of these values in terms of excellences? In other words, do we have reason to propose that educators offer ideas, images or visions of situations or traits of character that are deemed superb, excellent or perfect - or what we can call ‘ideals’ - as would be the implication of Waterman’s conception of the eudaimonic theory of human flourishing as I quoted in the second section? I cannot address this question to a full extent, but would like to propose two reasons as to why offering children values that are excellent within a genre or practice or excellent characteristics of person might be judged as a better version of good education. Firstly, following Raz I have already argued that children should learn about the standards of the genres of the objective goods. If one has an idea about what is the best within a genre, what would count as an excellence, one will probably evaluate current practices or alternative practices against these supreme standards. Ideas or images of for instance the trustworthy friend, the reliable colleague, the harmonious family or the just society function as regulative principles (Emmet, 1994) and may assist people in evaluating what they should do, change or continue to do in order to achieve these ideal standards. These ideas need to be those of excellences, which is the second reason, because they motivate people to strive for something higher or better than they might otherwise do (see also Rescher, 1987). Ideals can stimulate people to strive for a better life, a better society (for instance Jacoby, 2005). Heyting has critiqued my conception of ‘ideals’ as being not only reflective principles, but also unrealised ideas of excellence that a person hopes to 20
be able to realise and therefore pursuits. She conceptualises them as models or criteria ‘against which things in a relevant class can be assessed’” (2004, p. 242). I assume she would therefore agree with my first argument. Where we depart is that I believe that ideals are not only the criteria against which we assess a situation or a trait of character, but that they also have a pullfactor (see Rescher, 1987). Persons value the excellences to which the ideals refer so much that they are deeply motivated in trying to meet the criteria, even if they recognize that this is never completely possible.
Conclusion With this contribution I have given an example of the way in which philosophers of education can write about central topics in education. I hope to have shown that philosophy of education is an important discipline, because it aims to shed light on normative-ethical educational issues and to give arguments with which particular educational practices can be justified. I regret the fact that the importance attached to our discipline is in decline and that educational research is currently being dominated by the empirical paradigm. In my view academic education needs both empirical and theoretical or philosophical research in order to flourish. Empirical research offers us the necessary insights into educational practices or developmental processes - this contribution for instance depends quite strongly on empirical insights from psychology. Without these insights philosophers of education are in danger of becoming alienated from the practices about which they write. Empirical researchers, on the other hand, risk the danger of being so engrossed in data-collection and analysis that they neglect critical reflection on their premises and normative reflection on their results. Philosophers of education can ensure that such questions are not forgotten. I end with expressing the hope that when the time comes that my younger colleagues think that I deserve a symposium and book to commemorate my work, philosophy of education at Dutch universities is in a better state than it is today. My ideal is that there be flourishing departments at every university. Whether or not this ideal is realisable, I do not know, but that does not prevent me from trying to pursue it anyway.
21
References Aristotle (1985). Nicomachean ethics. (T. Irwin, Trans.). Indianapolis: Hacket Publishing Company. Baumeister, R.F. (1991). Meanings of life. New York: The Guildford Press. Baumeister, R.F. and Muraven, M. (1996). Identity as adaptation to social, cultural, and historical context, Journal of Adolescence, 19, 405-416. Callan, E. (1997). Creating citizens. Political education and liberal democracy. Oxford: Clarendon Press. Chirkov, V., Ryan, R. M., Kim, Y. & Kaplan U. (2003). Differentiating autonomy from individualism and independence: a self-determination theory perspective on internalization of cultural orientations and well-being, Journal of Personality and Social Psychology, 84 (1), 97-110. Deci, E. L. & Ryan, R. M. (1987). The support of autonomy and the control of behavior, Journal of Personality and Social Psychology, 53 (6), 1024-1037. Deci, A. L., Ryan, R. M. & Williams, G. C. (1996). Need satisfaction and the selfregulation of learning, Learning and Individual Differences, 8 (3), 165-183. De Ruyter, D. J. (2004). Pottering in the garden. On human flourishing and education, British Journal of Educational Studies 52 (4), 377389. Emmet, D. (1994). The role of the unrealisable. A study in regulative ideals. New York: St. Martin’s Press. Frankfurt, H. G. (1999). Necessity, volition and love (Cambridge, Cambridge University Press). Heyting, G. F. (2004). Beware of ideals in education, Journal of Philosophy of Education38 (2), 241-248. Jacoby, R. (2005). Picture imperfect. Utopian thought for and anti-utopian age. New York: Columbia University Press. Justitiële Verkenningen (2005). Kindbeelden. Den Haag: WODC. Kasser, T. & Ryan, R. M. (1993). A dark side of the American dream: Correlates of financial success as a central life aspiration, Journal of Personality and Social Psychology, 65 (2), 410-422. Murdoch, I. (1970). The sovereignty of the Good. London: Penguin. Peters, R. S. (1981). Ethics and education (9th impression). London: George Allen & Unwin ltd. Raz, J. (2003). The practice of value. Oxford: Oxford University Press. Reis, H. T., Sheldon, K. M. , Gable S. L., Roscoe, J. & Ryan, R. M. (2000). Daily well-being: The role of autonomy, competence and relatedness, PSPB, 26 (4), 419-435. Rescher, N. (1987) Ethical idealism. An inquiry into the nature and function of ideals. Berkeley: University of California Press. Ritzer, G. (2000). The McDonaldization of society (New Century Edition). Thousand Oaks: Pine Forge Press.
22
Ryan, R.M. & E.L. Deci (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being, American Psychologist, 55 (1), 68-78. Ryan, R.M. & E.L. Deci (2001). On happiness and human potentials: A review of research on hedonic and eudaimonic well-being, Annual Review of Psychology, 52, 141-166. Spiecker, B. & J.W. Steutel (2001). Reflective equilibrium as a method of philosophy of education, in: G. F. Heyting, J. White & D. Lenzen (eds.), Methods in philosophy of education. London: Routledge. Spiecker, B. & J.W. Steutel (2002). Sex between persons with ‘mental retardation’: An ethical evaluation, Journal of Moral Education, 31 (2), 155-169. Spiecker, B., J.W. Steutel & D.J. de Ruyter (2004). Self-concept and social integration. The Dutch case as an example, Theory and Research in Education, 2 (2), 161-175. Steutel, J.W. & B. Spiecker (2004). Cultivating sentimental dispositions through Aristotelian habituation, Journal of Philosophy of Education, 38 (4), 531-550. Van der Burg, W. en T. van Willigenburg (red.) (1998) Reflective equilibrium. Essays in honour of Robert Heeger. Dordrecht: Kluwer Academic Publishers. Waterink, J. (1958). Theorie der opvoeding. Kampen: Kok. Waterman, A.S. (1993). Two conceptions of happiness: contrasts of personal expressiveness (eudaimonia) and hedonic enjoyment, Journal of Personality and Social Psychology, 64 (4), 678-691. White, J. (2005). Educating for success. In: D.J. de Ruyter, G.D. Bertram-Troost & S.M.A. Sieckelinck (eds). Idealen, idolen en iconen van de pedagogiek. Amsterdam: SWP. White, J. (2005). Educating for success. In: D.J. de Ruyter, G.D. Bertram-Troost & S.M.A. Sieckelinck (eds). Idealen, idolen en iconen van de pedagogiek. Amsterdam: SWP. Wright, G.H., von (1963). The varieties of goodness. London: Routledge & Kegan Paul.
23
2 Searching for the good concerning education. Response to Doret de Ruyter CHRISTIANE THOMPSON
Which significance does the question of good education have within our academic discipline? The answer might not be as obvious as it might at first seem. Even though the question brings to language the requirements that everyday practice demands of educational science and thus, makes us focus on how to argue for reasonable means and ends of education, one could object that it narrows down our point of view regarding the phenomenon of education: By focusing on how education ought to be – is our analytic view not distracted from what is actually going on in everyday practice? Furthermore: Is it at all possible to pose the question of good education in a scientific setting – or does the inquiry into good education, rather, confront educational science with an insurmountable difficulty, because it demands answers that educational science cannot provide? As a matter of fact, the question of the good concerning education as well as its aims has often been set aside by educational researchers in the second the objective sciences, a considerable number of the efforts undertaken restricted themselves to descriptive and analytic issues. Normative aspects of education (as they were called) were ruled out because – so it was argued – they belong to a realm that does not allow concluding scientific results. This position, however, must remain unsatisfactory; for it implies a rift between scientific research and the life world – to use a term brought up by Edmund Husserl in the 1930s. In his ‘Crisis of the European Sciences and Transcendental Phenomenology’, Husserl came to the conclusion that the sciences took up a route that isolated them from the issues relevant to human existence. However, by turning away from these issues, the sciences lost their significance for culture, and thus found themselves in a deep crisis. Husserl urged the sciences to overcome their forgetfulness of the life world and reconsider the relationship to it. In a very similar fashion, one could argue that educational science 25
needs to be reminded of its significance for the life world – for human practice and culture; for everyday professional and non-professional educators have to think about educational issues and reach ‘good’ decisions in their practice. What could be their reference point if not one fostered by an educational scientific reflection? At this point, one can appreciate Doret de Ruyter’s contribution which clearly attempts to leave behind the problematic distinction of descriptive and normative dimensions of education, and which explicitly turns toward the question of good education from a scientific point of view. De Ruyter’s main approach lies in the development of the term ‘human flourishing’ that transfers the question of good education into the question of what is necessary for human well-being. Aside from those human needs that relate to the basic preservation of life, De Ruyter refers to psychological theories in order to conceptualize the term ‘well-being’. Following the SelfDetermination Theory by Ryan and Deci she argues that “there are three innate psychological needs that are essential to a person’s flourishing ... the need for autonomy or self-determination, for competence and effectance, and for relatedness or affiliation” (p.14). It is stated that these needs are universal even though they might express themselves differently, and are fulfilled in a variety of ways. Having determined human flourishing this way, De Ruyter launches the question of how it is possible to assist children in achieving their own well-being. It is here where values come into play, for they provide, according to De Ruyter, the framework for determining and interpreting what is good for oneself. Secondly, they grant the connection to social practice and its development. De Ruyter finally comes to an answer of her question: Education is good when “parents and professional educators pass on goods and values and teach children the capacities to reflect upon these to discover which conception of the good would contribute to their flourishing” (p. 20). Since I do not have the space here to adequately discuss De Ruyter’s line of argumentation, let me in the following focus critically on the term ‘human flourishing’ and thereby open a different perspective on the question of the good concerning education. This perspective will also allow me to say a few words regarding the conference topic, namely the relationship of different research generations. The difficulty that lies in the question of good education is how to understand and conceptualize ‘the good’. This difficulty is already present in Ancient Greek thought, e.g. in Plato’s dialogue ‘Protagoras’, where Socrates 26
asks about the ‘measure’ of the good. With respect to education: What could serve as a standard of quality for determining good education? De Ruyter approaches this question with the term ‘human flourishing’. I do not intend to offer an alternative conception. Rather, I would like to argue that our approach and handling of the question is significant for the question of good education itself. Let me explain what I mean by referring to a thought uttered by Heidegger in ‘Being and Time’: Whenever we raise a question (Heidegger concentrated on the question of Being), the question already bears references regarding its possible answers. By posing the question of good education in a determinate way, there are implications and presuppositions at work that shape our answer accordingly. Therefore, Heidegger turned toward the philosophical tradition in order to undertake its ‘destruction’. Here, destruction does not mean devastation or demolition but an analytic undertaking that makes apparent the structures that are at work in the question (of Being). In a similar fashion, I would like to analyse the moving forces in De Ruyter’s conceptual framework of human flourishing, and problematize its implications. The term ‘human flourishing’ obviously belongs to an anthropological framework. It is a term of fulfilment that gives education a developmental structure. The fulfilment refers to different needs that are thought to be specific to human existence in general. The complementing terms of need and fulfilment work on the basis of human agency, i.e. the modern conception of an autonomous and self-transparent subject. Thus, it comes not as a surprise that De Ruyter holds judgment to be significant for our relation to values. Education is supposed to bring about the ability to judge appropriately, and it is itself dependent on good judgment. To summarize, education is regarded as a supportive practice by adults toward children based on the modern anthropology of the knowing and active subject. It seems important to me to make these implications explicit, and wonder whether we could pose the question of good education differently. Could we understand education differently? Could we, e.g. understand it not as a supportive practice of responsible adults to not-yet responsible children, but as an interaction that confronts adults with the impossibility to completely understand the human and with the fact that their standpoints regarding lived values remain questionable and fragile? Could we see childhood not as a state that is to be overcome for the sake of development and flourishing, but as something of a higher dignity, something self-sufficient that – as Lyotard (2001) once suggested – might entail something more human than any other appearance of human existence? 27
I do not want to argue here for any position implied by these questions, but hope with the latter to convey that what we are, i.e. how we conceive of ourselves and our existence does not necessarily have to be the way it is. Foucault (1992) has called this procedure demarcating the borders of our existence, and he applied it in order to enable changes regarding our socially and historically determined existence. The historical and social formation of our existence also means that we have to reconsider De Ruyter’s statement that there are “innate psychological needs”. For even when admitting different types of interpretation and fulfilment of these needs, and thus seemingly allowing for plurality, De Ruyter’s line of argumentation is based on a specifically modern point of view: to speak of a psyche as an independent realm of being goes ultimately back to the Cartesian conception of two substances that exist independently from one another. The universalism becomes apparent once turning toward different cultures, where one discovers differing modes of selfunderstanding. Clearly, there is much more that can be discussed here. I hope we can take this point up in the following discussion as well as the idea of modern science that feels confident to be able to determine objectively and access empirically a subject-related state like ‘human flourishing’. What do these brief remarks imply for the question of good education? What I wanted to explicate is that we are always already determined regarding our idea of good education, even when only posing (scientifically) the question of good education. I believe it to be the central task of the philosophy of education to uncover the unthematized premises, and make our ideas and ideals that shape our very self-understanding accessible to us. It is here that the preceding research generations become important for us: When taking up the analytic view regarding unthematized premises, i.e. when following thinkers like Adorno (2003) and Arendt (2005) who proclaimed a thinking that is not content with itself but challenges itself and is essentially subversive, then the engagement with the tradition enriches our thought and grants us more possibilities in our theoretical and practical efforts. In this case, I hold the reference to the preceding research generations not as confining (in that it would determine our thoughts), but as actually granting us more freedom (because we encounter or uncover multiple interpretations of a problem, and can situate ourselves more responsibly). Following this line of thought, the accomplishments of the elder research generations will never be outdated. Rather, their questions and answers provoke new questions and answers. And it might be that this relationship is significant not only for the inquiry into good education but also for good education itself. 28
References Adorno, Theodor W. (2003): Anmerkungen zum philosophischen Denken. In Kulturkritik und Gesellschaft II. Eingriffe. Stichworte. Gesammelte Schriften: Band 10.2)(pp. 599-607). Edited by Rolf Tiedemann. Frankfurt/a.M.: Suhrkamp. Arendt, Hannah (2005): Denken ohne Geländer. München: Piper. Foucault, Michel (1992): Was ist Kritik? Berlin: Merve. Heidegger, Martin (1993): Sein und Zeit. Tübingen: Niemeyer. Husserl, Edmund (1996): Die Krisis der europäischen Wissenschaften und die transzendentale Phänomenologie. Eine Einleitung in die phänomenologische Philosophie. Hamburg: Meiner. Lyotard, François (2001): Das Inhumane. Plaudereien über die Zeit. Wien: Passagen. Plato (1990): Werke. Darmstadt: WBG
29
3 The changing structure of scholarly communication in educational science RAF VANDERSTRAETEN
Introduction In recent years, research into the social characteristics of science has taken an empirical and constructivist turn. The classical types of reflections on the institutional and normative embedding of science are being replaced by empirical research that focuses on how scientific knowledge is actually manufactured and shared with the wider scientific community. An important issue which has been brought to the fore in this context concerns the structure and evolution of scientific communication. As some authors have clarified, this issue provides a strong impetus for the analysis of scientific journal publishing. Communication in a scientific community is to a considerable extent dependent on journal publications. Analyses of the evolution of particular aspects of scientific journal publishing might therefore add to our understanding of the construction of the social identity of scientific disciplines. This paper presents a case-study which focuses on the evolution of scholarly publishing in education in the Netherlands. As in a number of other countries, the institutionalisation of educational science in the Netherlands took place in the course of the 20th century. Often the official beginning of the discipline in the Netherlands is marked as the 3rd February 3 of the year 1900, when the first lecturer in education (J.H. Gunning) delivered his inaugural address at the University of Utrecht. But it is not the determination of a point zero that is currently of interest. On the basis of an analysis of the journal Paedagogische Studiën [Studies in Education, henceforth PS], I aim to examine the changing structure of communication in this discipline. The journal PS lends itself very well to this kind of research. Founded immediately after the First World War, it is the first central disciplinary journal in the field of the leading role in the field. It has covered a broad range of topics characteristic of the field of education, and has continued uninterrupted 31
education with forthright scientific pretensions. Between the 1920s and 1970s, the journal has continued to play periodic publication for several decades.1 On the basis of an analysis of the PS journal during the period of 1920 to 1970, it is in my view possible to analyse the changing patterns of communication in this discipline in the Netherlands. My purpose is to highlight and analyse the role played by specialized journals in the formation of academic fields of study. Such journals carry, channel, and give shape to the fields’ communicative processes. They do so in ways that have far-reaching consequences for the social recognition of relevant topics, legitimate issues, and methodological standards. They also do so in ways that pre-structure how different generations can contribute to the development of the scientific discipline.2 Hereafter, I first briefly explain the core theoretical notions that underpin my approach. Afterwards, I present and discuss statistical data that represent the evolutions that have taken place in the communicative structures of Dutch studies in education. In the concluding section, I summarise my findings and point to some effects that concern the relation between different generations of educational researchers.
Disciplines as communication systems From the 19th century onward, the university has become the prime location where scientific research takes place, where new generations of researchers can be trained and recruited, and where scientific careers can be pursued. The unquestionable relationship between the growth of the university, on the one hand, and of scientific research, on the other, has led several researchers to conceive the social history of science in terms of the history of chairs and professorships at universities or research institutes. However, these institutes are not the locations where research findings are communicated, published 1
From the 1970s onwards, the journal has clearly evolved into a subdisciplinary journal, specializing in school psychology. But the journal was/is well aware of its leading role in the field. It has ‘celebrated’ its own history on several occasions (1961: 273-276; 1974, nr. 1; 1988: 502-513; 1998, nr. 6). 2 In the same way, one can ‘read’ the social networks, figurations or systems within which education takes place as networks or systems which determine how education is and is not possible, and how education actually takes place. This is, in my view, also the research approach which Frieda Heyting has propagated and used in a number of her writings (e.g. Heyting, 1987, 1997). Its applicability across a wide range of themes might be perceived as an indication of its strength.
32
and evaluated. The communication of research findings takes places in another context, namely, a disciplinary context. Research findings are communicated via encyclopaedias and books, or at conferences, seminars and workshops. But most of all, the scholarly journal has become the instrument that allows for the self-organization of the disciplines. Publications in a scholarly journal (articles, reviews, rejoinders, etc.) have become the basic communicative units in the discipline (Stichweh, 1994, p. 52-83). They generate and reproduce the disciplinary communication process. Each publication interacts with preceding ones, by incorporating into its own line of reasoning arguments developed in other publications; and each new publication, due to the claims it makes to new knowledge, invites reactions and hence further publications. It is important to note that scholarly journals not only enable the communication of research findings, but also influence how contributions to scientific communication can be made.3 In comparison with the production and circulation of books, periodicals lead to the rapid succession of small contributions. Publications in periodicals succeed each other at short and regular intervals. Furthermore, the scientific review is likely to represent the theoretical and methodological variability of a (sub)discipline, as it consists of a collection of different articles by different authors. It can be expected that the diversity of the published contributions will provoke reflection upon the relationship between, and the coherence of these contributions. Also, readers who subscribe to a scientific journal do not know in advance what issues will be raised in what ways in that journal. A journal thus evokes expectations which are different from the ones evoked by a scientific book. A journal is expected to present a picture of the state-of-the-art of an entire field of inquiry. Seen against this background, the idea seems to suggest itself to analyse scholarly journals with regard to discerning the very constitution of scientific disciplines as they have developed in the past decades (Tenorth, 1990; Keiner, 1999). Periodicals influence the temporal structure of the system of science (Vanderstraeten, 2006). The periodicity of appearance presses scientists to publish at regular intervals (‘publish or perish’). The findings of a journal article can already be superseded in the next issue of the journal. At the same time, journals and their editorial boards can regulate and control access to 3
For the history of scientific journals, see Smith (1972), Kronick (1976), McKie (1979), Manten (1980a), Bazerman (1988), Atkinson (1999). Most of these journals evolved from the newsletters of scientific societies. For many years, journals played a secondary role in relation to books.
33
scientific communication. Multiple norms and values, which have been developed within the system of science, directly bear upon publication behaviour. Discussions about what is, and what is not, a valid publication are ageold (Manten, 1980b). The current peer review system puts up a barrier, but at the same time grants a minimal form of recognition or credit to published research findings. It endorses what the scientific community takes to be certified knowledge (Garfield, 1985). The scientometric instruments that have been developed in the past decades - such as Journal Citation Reports and Journal Performance Indicators - have strengthened the relevance of periodicals. One may therefore conclude that the scholarly journals play a prominent role in the formation of scientific disciplines. The analysis of the foundation and evolution of a scholarly journal allows us to examine the evolution of a (sub)discipline, as well as the specific position of the journal within the field of study. The weight I give to scholarly journals may be somewhat surprising. As I have indicated, disciplines would scarcely be able to survive without the support of institutions - be these universities, academies, research institutes, or any other form of organizational infrastructure - since they guarantee the continuity of academic work by providing occupational roles, social status, and publication facilities. Scientific disciplines, in addition, have to rely on delimited groups of people who join together on the basis of common interests or motives, especially within the frame of conferences, study groups, professional associations, or scholarly societies (Titze, 1995; Horn, 2003). I do not doubt that overviews of this ‘infrastructure’ are helpful in order to understand the evolution of the system of science, and of particular disciplines. But if one conceives of scientific disciplines as self-organizing networks of communication, then it becomes necessary to analyse how the flow of a disciplinary communication process is generated, continued, and reproduced. I have indicated that publications stand for the continuity of a discipline’s communication process, and that specialized journals are media of publication par excellence. There are good reasons, therefore, to analyse specialized journals in order to discern the distinctively patterned networks of communication that characterize scientific disciplines. There are also good reasons to study how structural characteristics influence how generations may contribute to the development of the discipline (here: education) and how they may constitute a difference that makes a difference.
34
Educational science in the Netherlands How does one proceed with such an analysis? In the introduction of his excellent study on the history of the American Journal of Sociology (founded in 1895), Andrew Abbott remarks: ‘I was plowing what seemed virgin soil; there was almost no serious historical investigation of the institutional structures of modern scholarly publication’ (1999, p. ix-x; see also Abbott, 2001, p. 91-120; Platt, 2002). In contrast to past empirical studies of scholarly journals which have mostly focused on the effects of an author’s rank, method, gender or institutional affiliation on the publication of particular articles in particular journals (Hammermesh, 1994; Hirschauer, 2004), I try to discern the constitution of scholarly communication in the field of education as it has developed in the Netherlands. My predominantly quantitative analyses of the following subsections focus on the changing role of the editorial board of PS and on the uses or citations of publications in journal articles in the period between 1920 and 1970 (or 1975, if this provides a better picture of a particular trend).4 It is not possible to provide a simple explanation of every trend, but taken together these data elucidate the changing role of the flagship journal in the changing communication system of educational science in the Low Countries. Authors and editors From the onset, PS conceived itself as a scholarly journal.5 Most of the founding editors held research positions at Dutch universities. The board has always been chaired by noted academics: J.H. Gunning (1919-1938), Ph. Kohnstamm (1938-1951), I.C. van Houte (1952), H. Nieuwenhuis (19531957), Ph. Idenburg (1958-1972). At the end of the 1960s, the increasing specialization of research and the increasing impact of hypothesis-testing 4
I acknowledge the help of Ivo van Hilvoorde, who assembled the statistical data. For a previous interpretation of these data, see Vanderstraeten & van Hilvoorde (2001). 5 Around 1900, several educational journals with scientific pretensions were founded, viz. Nieuw Tijdschrift ter Bevordering van de Studie der Paedagogiek [New Journal for the Advancement of the Study of Education, 1890-1908], Oud en Nieuw [Old and New, 1896-1902], Nieuwe Paedagogische Bijdragen [New Pedagogical Contributions, 1901-1906] and Kinderstudies [Child Studies, 1916-1922]. But these journals served as the mouthpiece of the interests prevailing in particular teacher organizations, research institutes or religious groups. Moreover, for various reasons (lack of subscriptions, lack of contributions, conflicts among board members), all of them soon disappeared. Partly due to its institutional embedding in the public universities of the Netherlands, PS established itself as the flagship journal of the discipline in the Netherlands.
35
and empirical-analytical thinking left their mark on the landscape of scholarly journals in the Low Countries. The broad, generalist character of PS was subjected to great pressure (cf. PS, 1988, p. 502-513; PS, 1998, nr. 6, p. 944). In 1970, the journal slightly changed the spelling of its name - from the ‘archaic’ Paedagogische Studiën to the ‘modern’ Pedagogische Studiën. In this period, it also started to narrow its focus both substantively and methodologically, specializing in (quasi-)experimental articles on school psychology. With this background in mind, the following analyses are based on publications in the period until 1970. To highlight particular processes of scholarly production, figure 1 focuses on the role of the editorial board of PS in the period 1920-75. This figure displays the proportion of articles in PS which were written by members of the editorial board. Although there are some fluctuations, it can readily be seen that the role of the editorial board was redefined in the 1960s. Around 1950, the editorial board still authored almost half of all published articles.
0,8
60
0,7
50
0,6 40
0,5 0,4
30
0,3
20
0,2 10
0,1 1974
1971
1968
1965
1962
1959
1956
1953
1950
1947
1944
1941
1938
1935
1932
1929
1926
1923
0 1920
0
total amount of articles
ratio of articles editorial board
Figure 1: The productivity of the editorial board
In the 1970s, this proportion dropped to about ten percent of the journal contributions. Thus, the editorial board of PS came to play a less visible role in the scientific scene during this period, especially in comparison with the years before the Second World War. This evolution has to do with the fact that an increasing number of authors became less loyal to PS and wanted to 36
publish in several journals. The rise of competing Dutch periodicals and the growing internationalisation of educational science in the Netherlands contributed to this trend. Moreover, publication in PS became less important to the editors’ own academic progress. The number of full professors on the editorial board of PS increased steadily (in 1936, 22%; in 1946, 38%; in 1956, 44%; in 1966, 59%; in 1976, 63%; in 1986, 70%). It was a new and relatively large group of younger researchers that used the journal as a publication outlet for its work. The role of the editorial board changed accordingly. Instead of filling the pages of the journal with their own contributions, the members of the editorial board became increasingly engaged as gatekeepers of scientific communication channels (cf. Gieryn, 1999). To complement the preceding analysis, table 1 provides an overview of the distribution of publications in PS. My starting point was the following question: Does a select group write most articles or is the majority written by a large group of authors who publish only once or twice in PS? The first column of this table lists the number of articles written by individual authors; the second column lists the corresponding number of authors. Column 3 and 4 display the evolution in terms of percentages. The names of the most productive authors of PS are given in column 5. An asterisk identifies the members of the editorial board among these productive authors. The last columns show the total amount of articles (in absolute numbers and percentages respectively). For example, there are two authors (0.3%) who published 25 articles in the period 1920-1970 (Van Eck and Van Veen); 98.6% of the authors published fewer than 25 articles each, but as a group only contributed 75.4% of the total number of articles published in PS (in absolute numbers: 1169 out of 1550 articles). Table 1 shows that 64.4% of the authors published only one article in PS between 1920 and 1970. On the other hand, 10% of the authors wrote more than half of the total amount of contributions. The eight most productive authors - all of whom were important members of the editorial board - together published almost 25% of all articles. The top three (Van der Velde, Langeveld en Kohnstamm) were responsible for 12.6% of all articles. In this period, PS is thus characterized by an oligarchic pattern of scientific communication.6 Until the end of the Second World War, a limited number of 6
These percentages decrease drastically when the period after 1970 is taken into account. On the other hand, the dominance of a small number of authors was even greater in the pre-War period. Between 1920 and 1941, three highly productive writers – Gunning (47), Diels (47) and Kohnstamm (59) – wrote 24.2% of all the articles, and the ‘top 10’ wrote 41.5% of the total amount. Such proportions are not unique to educational science in the Netherlands.
37
Table 1: The productivity of the authors Arti- Authors % Cum. cles % 1 376 64,4 64,4 2 99 16,9 81,3 3 36 6,2 87,5 4 19 3,2 90,7 5 14 2,4 93,1 6 6 1,0 94,2 7 4 0,7 94,9 8 6 1,0 95,9 9 1 0,2 96,1 10 3 0,5 96,6 11 3 0,5 97,1 12 2 0,3 97,4 14 1 0,2 97,6 15 2 0,3 97,9 Brugmans*; Siewertsz v Reesema 16 1 0,2 98,1 L. van Gelder* 23 1 0,2 98,3 Ph. Idenburg* 25 2 0,3 98,6 P.L. van Eck; G. van Veen* 26 1 0,2 98,8 H. Nieuwenhuis* 27 1 0,2 99,0 H. Stellwag* 37 1 0,2 99,1 P. Post* 47 1 0,2 99,3 P. Diels* 48 1 0,2 99,5 J.H. Gunning* 60 1 0,2 99,7 I. van der Velde* 63 1 0,2 99,8 M.J. Langeveld* 73 1 0,2 100,.0 Ph. Kohnstamm* Total
548
Total Cum % articles of total 376 24,3 574 37,0 682 44,0 758 48,9 828 53,4 864 55,7 892 57,5 940 60,6 949 61,2 979 63,2 1012 65,3 1036 66,8 1050 67,7 1080 69,7 1096 70,7 1119 72,2 1169 75,4 1195 77,1 1222 78,8 1259 81,2 1306 84,3 1354 87,3 1414 91,2 1477 95,3 1550 100.0
100%
Note: An asterisk identifies the editors among the most productive authors.
individual editors and authors unmistakably left their mark on PS and on educational research in the Netherlands. But from the 1960s onwards, the discipline and the journal became less dependent on dominant figures (with their particular individual interests). Instead, the work of a larger scientific
Baumert & Roeder (1990), for example, found nearly identical ratios about the productivity of German professors in physics and in education.
38
community came to the fore in the journal. The organization of scientific communication increasingly took place by means of journal publications. Publications and citations in PS I have indicated that the communicative network of a discipline consists of publications that do (or do not) refer to each other. These references to other publications reveal the collective character of the scientific practice; the identity of a discipline can be established by way of citations. New findings are linked back to already published results (albeit often in a chaotic way). It is this structure that makes publications elementary units of the system of science (Stichweh, 1994, p. 52-83). Conversely, citation analysis provides insight into the communication structure of a discipline or a scientific periodical. I have made both a quantitative and a content analysis of PS between 1920-1970. The following questions underlie my analysis: What is the number of references to periodicals and books in PS-articles, and how does this relationship evolve? Is there an evolution in the use of German and English publications (as has been observed with regard to the evolution of psychology in the Netherlands)? Familiarity with often-quoted sources in part determines the readership’s identification with a journal. It leads to the development of social and intellectual boundaries between disciplines. One also speaks in this respect of the “geography” of the discussion forum of scientists. Figure 2 analyses the origins of citations. This figure clearly indicates that the discussion forum for PS has been dominated by Dutch sources. In comparison with trends in the field of psychology, the fairly parallel evolution of references to English and German publications until the midst of the 1960s is striking (van Strien, 1993, p. 158). The late breakthrough of references to English literature is probably related to the lasting influence of the ‘geisteswissenschaftliche’ tradition in education. On the other hand, the ‘rate of circulation’ of sources significantly increased in the course of this period. The books by dominant figures such as Kohnstamm and Langeveld did survive relatively long. In the 1950s and 1960s, a few books of foreign origin were frequently quoted, such as Schelsky’s Die skeptische Generation (1957) and Bloom’s Taxonomy of Educational Objectives (1956/1964). However, such cornerstones remained exceptional in the building of educational science (De Solla Price, 1963).
39
Figure 2: Language of citations 300 250 200
Dutch German
150
English
100
French
50
1970
1968
1966
1964
1962
1960
1958
1956
1954
1952
1950
1948
1946
0
The genesis of the contemporary, standardized system of citations and references (APA-style) did not follow a linear pattern in PS. Although bibliographies and lists of cited literature already existed in the 1920s, it was not until the 1960s that they became widely used (Leydesdorff, 1998; Wouters, 1999; Moed, 2005). Until then, it was common to use footnotes or to omit references altogether. In 1959, an author could still write: ‘the following does not claim to be original. A lot is borrowed from other publications, but without reference’ (PS, 1959, p. 87). Even in the 1960s, acknowledgement of sources was not yet obvious. ‘It goes for the whole article that the inclusion of references to other authors is impossible, as these notes would be more extensive than the article itself’ (PS, 1966, p. 125). Finally, an almost complete standardization took place around 1969/1970. Not completely by coincidence, it was also in 1970 that the style and the layout of PS changed. The journal got a ‘modern’ look. In the meantime, the number of references had drastically increased (to almost ten times the number of 1946), with the most common reference being to Dutch publications. Figure 3 offers an overview of the type of sources - books or journal publications - that were listed in bibliographies and footnotes. The category ‘books’ also includes governmental reports. The remarkable dominance of ‘book’ publications is a consequence of the profile of PS. The journal used to pay a lot of attention to contemporary issues and policy documents. The increase of the number of references to periodicals was not proportionate to 40
that of books. Even the number of ‘self-citations’ - that is, citations to other publications in PS - remained remarkably low. Despite frequent positive reviews in PS of foreign periodicals, German and English journals continued to play a marginal role in the disciplinary communication in the Netherlands. Between 1920 and 1970, PS was primarily a journal that compiled and published information which was not readily available to its subscribers. It also stimulated critical reflection on prevailing educational policies. The role PS fulfilled in this period was mainly one of reporting and indicating developments in the field of education and educational science (as the high number of announcements and reports of conferences also demonstrates). Figure 3: Citations from books and journals 450 400 350 300 250 200 150 100 50 0
journals
1970
1968
1966
1964
1962
1960
1958
1956
1954
1952
1950
1948
1946
books
Finally, figure 4 displays the evolution of the number of authors or coauthors per published article. Before 1941, a publication with more than one author was exceptional. There is, on average, not more than one co-authored contribution per annual in this period. Afterwards, the authorship of publications is increasingly shared with one or even more colleagues. Besides wellknown strategic considerations, research-intrinsic developments also influenced this evolution - as empirical research is often carried out in teams. This evolution is, of course, not solely a Dutch phenomenon. The similarity to authorship trends in German periodicals (especially the Zeitschrift für Pädagogik) is remarkable (Keiner, 1999). Compared to developments in the field of scientific psychology, however, this evolution occurred relatively 41
late (i.e. not in the fifties but only at the end of the sixties). As indicated, the analysis of citations pointed to a somewhat similar time lag. German researchers refer to this as the ‘scientific retardation’ of the discipline of education (Tenorth, 1989). Figure 4: Number of authors per article 60 50 40
5 authors 4 authors
30
3 authors 2 authors
20
1 author
10
1974
1972
1970
1968
1966
1964
1962
1960
1958
1956
1954
1952
1950
1948
1946
0
Concluding remarks From the 19th century onwards, it makes sense to describe scientific communities as communication networks that heavily rely on specialized journals for their own generation, continuation, and reproduction. The scientific journals represent the communicative form by which, at the macro-level of the system of science, communication complexes specialized along disciplinary lines can be bound together and persist in the long run. Moreover, the scientific publication affects the way research is conducted; it interferes in the scientific ‘production process’. In a kind of feedback loop, publications exercise pressure on the scientific production process, and thus contribute to the integration and identity of scientific disciplines. 42
In the first decades of the 20th century, the field of education in the Netherlands was dominated by the generation of its academic founding fathers. After the Second World War, it gradually became clear that the field could no longer identify itself with certain prominent figure-heads (with their idiosyncratic interests). An increasing number of researchers started to contribute to scholarly discussions. In this context, publications in educational journals came to play a prominent role in the development of disciplinary communication structures. The graphs and tables presented in this paper indicate how the major specialized journal in the Netherlands, Paedagogische Studien, has fulfilled its role in the period 1920-1970 (1975). There is no doubt that the purposes of this journal have changed significantly over the course of this period. For a long period of time, the editorial board tried to disseminate findings of international research to researchers in the Netherlands, and to offer its readers an overview of national and international developments in education. The board was itself largely responsible for writing the articles of PS. This kind of informative role has gradually disappeared in the second part of the 20th century. At present, the raison d’être of the journal is the presentation of new research findings and new insights. Its readership now consists of potential authors of journal articles (who do not need a Dutch journal to learn about new, international developments in their field). It is an important indication of the fact that the history of science is characterized by a shift of the meaning of ‘discipline’, namely, from an imperative to preserve the truth to an interest in the novelty of an invention.7 What is communicated might be a small particle of knowledge, as long as it is a new particle of knowledge. A contemporary discipline, such as education, is based on the incessant production of novelties. In the second part of the paper, I have also pointed to an increase in coauthorship of articles. For PS, this trend began in the 1960s. The indexes of recent volumes of this journal (and of many other scholarly journals, too) abundantly illustrate that this trend has become more outspoken during the last decades. Moreover, the current rise of ‘virtual’ journals tends to rein7
The term ‘discipline’ is derived from the Latin discere (learning); disciplina has long been used as a term for the ordering of knowledge for the purposes of instruction in schools and universities. The term disciplina also included implications such as admonition, correction and even punishment for mistakes (Foucault 1984). In the early modern developments, the archival function of disciplines still dominated (Rorty 1979: 131-139; Stichweh 2001). The discipline was a place where one deposited knowledge after having found it out, but it was not an active system for the production of knowledge. It is only in the course of the 19th and 20th century that the disciplinary structures acquire a much more determining role in the system of science.
43
force this trend. In fields such as physics, biology, mathematics, or information sciences, scholarly articles written by only one author have become highly exceptional (Zwart, 2001, p. 26-29). The increase in co-authorship goes hand in hand with the further standardization of the methodology, terminology, and composition which are used in scholarly publications. In line with Michel Foucault (1995, p. 789-809), we might therefore speak of the ‘disappearance’ and ‘effacement’ of the author; the ‘subjectivity’ of the author tends to get lost in specialized scholarly publications. Perhaps it is one of the contradictions of the postmodern society that publications in scholarly journals have in recent years at the same time become more important for purposes of evaluating individual researchers and research groups. The modern system of science has developed its own social dynamics. Scientific disciplines are communication systems; their analysis requires an analysis of the diverse characteristics of their communication patterns. As we have seen, the changing structure of scientific communication makes knowledge claims less dependent on the person or authority of individual authors. (Post)modern science is not based on the achievements of extraordinary individuals but on the epistemic force of both national and global disciplinary communities. At present, there exist structural pressures which call for regular and frequent publications, both of members of the new and the older generations. There also exist structural pressures away from singleauthored and towards co-authored publications - for example, between Ph.D. students and their supervisors, or between other junior and senior members of research teams. In the course of the 20th century, the social conditions of scholarly publishing have changed in important ways. In my view, it can therefore be maintained that the different generations do not (or do no longer) carry the specific knowledge claims of different scientific disciplines - in the way which Karl Mannheim (1928) first described and analyzed.
References Abbott, A. (1999). Department & discipline. Chicago: University of Chicago Press. Abbott, A. (2001). Chaos of disciplines. Chicago: University of Chicago Press. Atkinson, D. (1999). Scientific discourse in sociohistorical context. The philosophical transactions of the Royal Society of London, 1675-1975. Mahwah: Lawrence Erlbaum Associates.
44
Baumert, J. & Roeder, P. M. (1990). Forschungsproduktivität und ihre institutionellen Bedingungen - Alltag erziehungswissenschaftlicher Forschung. Zeitschrift für Pädagogik, 36, 73-97. Bazerman, C. (1988). Shaping written knowledge. The genre and activity of the experimental article in science. Madison: University of Wisconsin Press. Bloom, B. S. et al. (1956-1964). Taxonomy of educational objectives. New York: Longman/MacKay. De Solla Price, D. J. (1963). Little science, big science. New York: Columbia University Press. Eisenstein, E. (1979). The printing press as an agent of change. Cambridge: Cambridge University Press. Foucault, M. (1984). Surveiller et punir: Naissance de la prison. Paris: Gallimard. Foucault, M. (1995). Qu’est-ce qu’un auteur? In: Dits et écrits I. Paris: Gallimard. Garfield, E. (1985). The awards of science and other essays. Philadelphia: ISI Press. Gieryn, T. F. (1995). Boundaries of science. In: S. Jasanoff, G. Markle, J. Petersen & T. Pinch (Eds.), Handbook of science and technology studies. Thousand Oaks: Sage. Gieryn, T.F. (1999). Cultural boundaries of science. Credibility on the line. Chicago: University of Chicago Press. Hammermesh, D. S. (1994). Facts and myths about refereeing. Journal of Economic Perspectives, 8, 153-163. Heyting, G. F. (1987). Autonomie en socialiteit in de opvoeding. Leuven: Acco. Heyting, G. F. (1997). Het vanzelfsprekende en het discutabele. Een schets van opvoedkundig grondslagenonderzoek. Utrecht: SWP. Hirschauer, S. (2004). Peer Review Verfahren auf dem Prüfstand. Zeitschrift für Soziologie, 33, 62-83. Horn, K. -P. (2003). Erziehungswissenschaft in Deutschland im 20. Jahrhundert. Bad Heilbrunn: Klinkhardt. Keiner, E. (1999). Erziehungswissenschaft 1947-1990: Eine empirische und Vergleichende Untersuchung zur kommunikativen Praxis einer Disziplin. Weinheim: Deutscher Studien Verlag. Kronick, D. A. (1976). A history of scientific and technical periodicals. The origins and development of the scientific and technical press, 1665-1790. Metuchen: Scarecrow Press. Leydesdorff, L. (1998). Theories of citation? Scientometrics, 43, 5-25. Luhmann, N. (1990). Die Wissenschaft der Gesellschaft. Frankfurt a.M.: Suhrkamp. Mannheim, K. (1928). Das Problem der Generationen. Kölner Vierteljahreshefte für Soziologie, 7. Manten, A. A. (1980a). The growth of European scientific journal publishing before 1850. In: A.J. Meadows (Ed.) Development of science publishing in Europe. Amsterdam: Elsevier. Manten, A. A. (1980b). Publication of scientific information is not identical with communication. Scientometrics, 2, 303-308.
45
McKie, D. (1979). The scientific periodical from 1665 to 1798. In: A.J. Meadows (Ed.) The Scientific Journal. London: Aslib. Moed, H. F. (2005). Citation analysis in research evaluation. New York: Springer. Platt, J. (2002). The history of the British Sociological Association. International Sociology, 17, 179-198. Reill, P. H. (1998). The construction of the social sciences in late eighteenth and early nineteenth century Germany. In: J. Heilbron, L. Magnusson & B. Wittrock (Eds.) The rise of the social sciences and the formation of modernity. Dordrecht: Kluwer Academic Publishers. Rorty, R. (1979). Philosophy and the mirror of nature. Princeton: Princeton University Press. Schelsky, H. (1957). Die skeptische Generation. Düsseldorf: Diederichs. Smith, A. G. R. (1972). Science and society in the sixteenth and seventeenth centuries. London: Thames & Hudson. Stichweh, R. (1994). Wissenschaft, Universität, Professionen: Soziologische Analysen. Frankfurt a.M.: Suhrkamp. Stichweh, R. (2001). History of scientific disciplines. In: N.J. Smelser & P.B. Baltes (Eds.) International Encyclopedia of the Social & Behavioral Sciences. Amsterdam: Pergamon. Storer, N. & Parsons, T. (1968). The disciplines as a differentiating force. In: E.B. Montgomery (Ed.), The foundations of access to knowledge. Syracuse: Syracuse University Press. Tenorth, H. -E. (1989). Deutsche Erziehungswissenschaft im frühen 20. Jahrhundert. Aspekte ihrer historisch-sozialen Konstitution. In: P. Zedler & E. König (Eds.), Rekonstruktionen pädagogischer Wissenschaftsgeschichte. Fallstudien, Ansätze, Perspektiven. Weinheim: Beltz. Tenorth, H. -E. (1990). Vermessung der Erziehungswissenschaft. Zeitschrift für Pädagogik, 36, 15-27. Titze, H. (1995). Wachstum und Differenzierung der deutschen Universitäten 18301945. Göttingen: Vandenhoeck & Ruprecht. Vanderstraeten, R. & Hilvoorde, I. van (2001). Evolutielijnen van de wetenschappelijke pedagogiek: disciplinegrenzen, tijdschriftpublicaties en Paedagogische Studiën. Pedagogische Studiën, 78, 36-55. Vanderstraeten, R. (2006). The historical triangulation of education, politics and economy. Sociology, 40,125-142. Van Strien, P. (1993). Nederlandse psychologen en hun publiek. Assen: van Gorcum. Wouters, P. F. (1999). The citation culture. Amsterdam: n.p. Zwart, H. (2001). De wetenschapper als auteur. Nijmegen: SUN.
46
4 The discontinuous development of scientific disciplines: Some remarks on scholarly authorship. Response to Raf Vanderstraeten ROEL VAN GOOR
With the presentation of the results of an empirical analysis of the prominent Dutch educational journal Paedagogische Studiën [PS] Vanderstraeten aims to examine the changing structure of communication within education as a scientific discipline. It may seem surprising to come across a paper that presents empirical research in a collection that has a strong philosophical orientation. However, ever since Quine naturalized epistemology and Kuhn historicized the image of science the empirical study of scientific discourse increasingly has been acknowledged as a fruitful source for philosophical reflection1. Against this background Vanderstraeten’s research that focuses on “how scientific knowledge is actually manufactured and shared with the wider scientific community” (p. 31) can be considered a useful and welcome contribution to philosophy of education. I would like to start by pointing out that I can go along with Vanderstraeten’s point of departure that we can conceive of scientific (sub)disciplines as communication systems with publications in journals as their basic units of communication. I can also go along with the idea that the collection of publications in a journal gives us an insight into a discipline’s central characteristics. He states that: “the scientific review is likely to represent the theoretical and methodological variability of a (sub)discipline” (p. 32). Against this theoretical background the choice of analysing the changing structure of communication of PS in order to get a grip on transformations in education as a scientific discipline is understandable and promising. Based on his analysis Vanderstraeten concludes that the modern system of science has developed a social dynamics in which the subjectivity of indi1
Heyting, Hemrica, Lepková en Mulder for instance, have convincingly demonstrated the relevance of the analysis of discourse for philosophical reflection on education and upbringing (Hemrica & Heyting, 2004; Heyting, 2001; Lepková et al., 2004).
47
vidual authors tends to eventually disappear. This conclusion might seem intelligible in the light of the current state of scholarly communication in education. I do think, however, that this conclusion puts too much emphasis on the disciplining force of structures of communication, and that it fails to explain the discontinuous historical development of scientific disciplines. Vanderstraeten stresses that the evolution of scholarly communication has to be understood in the light of the interrelatedness of journal publications. “Each publication interacts with preceding ones, by incorporating in its own line of reasoning arguments developed in other publications; and each new publication, due to the claims it makes to new knowledge, invites reactions and hence further publications” (p. 33). As a consequence of this process publications do not only represent a progressive, but also a conservative force within disciplines. “They generate and reproduce the disciplinary communication [italics added]” (p. 33). Although I agree with Vanderstraeten’s analysis that the influence of publications on the communicative process has two dimensions, it seems to me he puts too much weight on the reproductive dimension. The ‘feedback loop’ – as he calls it – of publications contributes in his view to the “integration and identity of scientific disciplines” (p. 43). Vanderstraeten seems to presuppose a process of convergence towards an increasing consensus on the rules of conduct – for instance norms for procedures of justification, terminology, acceptability of methodological instruments or types of research questions – within a scientific discipline. Of course, most of the time communication processes proceed without much noise. Indeed, in those periods every journal publication only adds “a new particle of knowledge” (p. 43), leaving the more general norms and rules concerning the nature and production of knowledge that guide the communication untouched. But to my view major transformations can take place over a relatively short period in time. Kuhn (1962)’s The structure of scientific revolutions has convincingly shown that scientific progress does not (only) proceed gradually, but in fits and starts. His ideas on scientific revolutions teach us that the development of scientific disciplines can consist in a replacement of some of a discipline’s most basic presuppositions, leaving the discipline almost unrecognizable. Looking at the data presented to us by Vanderstraeten, it seems to me that a major transformation of the educational discipline has actually taken place in the period that his analysis focuses on. Quite suddenly from the midst of the 1960s significant changes occur in the structure of communication of PS. Vanderstraeten reports on an increas48
ing standardization of methodology, the genesis of a standardized system of citations and references, and a breakthrough of references to English literature leading to a climax around 1970. At that point, Vanderstraeten notices, PS started to narrow its focus both substantively and methodologically, specializing in (quasi-)experimental articles on school psychology” (p. 36). These findings seem to indicate a radical transformation of education as a scientific discipline. The discipline seems to have shifted towards a more quantitative-empirical model of social science in general, and cognitivist psychology in particular, leaving behind its Geisteswissenschaftliche roots. These findings seem to be at odds with the idea of a continuous convergence of the journal’s communicative structure. Furthermore, they seem to contradict the conclusion that the subjectivity of the authors of journal publications tends to get lost in modern science. If journal publications account for the formation and transformation of scholarly communication, than the authors that published in PS in the 1960s did not disappear from the scientific stage. On the contrary, they seem to have made an overwhelming appearance as revolutionaries. Apparently the authors evoked a revolutionary transformation of education as a scientific discipline. Did Vanderstraeten fail to notice a revolution? Obviously, he did notice a major shift had occurred. But what could have led him to ignore the influence of the authors in this process of radical transformation? I think Vanderstraeten has not been able to explain the authors' influence because of his – Foucauldian – conception of disciplinary development. Vanderstraeten follows Foucault in his analysis by arguing that within the modern system of science authors are increasingly constrained by disciplinary structures (p. 43n. 7). Participating in scientific discourse means subjugation to its implicit norms; if an author won’t act according to disciplinary rules, he will be corrected, punished, and be in danger of being excluded from the communication (cf. p. 43n. 7). In this interpretation the historically generated powers immanent in institutionalized communication systems work through the individual, though not being under his control. This indicates why Foucault speaks of disciplines in terms of “regimes” (Foucault, 1975). Scholarly authorship then is like becoming bound up in an entangling knot that tightens with every article that is published. To my view, in this interpretation the possible contribution authors can make to communication systems is heavily underestimated. Vanderstraeten conceives of authorship as located within communication systems with authors only being able to take small steps within the margins given by the norms underlying the system at hand. I would like to suggest an 49
interpretation of authorship as acting upon a communication system, trying to bring about – either smaller, of larger – transformations in the communicative process. Of course, being intelligible still requires subjugation to norms. The difference is, however, that a system’s norms do not signify a limit that cannot be surpassed, because any norm, any rule of the game that is being played may be brought up for discussion at any moment (cf. e.g. Brandom, 2000; and Williams, 2001). Williams argues in this respect that it is not only possible to raise the discipline’s ‘level of scrutiny’, but that one might also change its ‘angle of scrutiny’ (Williams, 2001, p. 227). This would give an author the opportunity to try to transform a discipline in an unforeseeable manner. One might argue against this position that articles that aim to challenge basic taken-for-granted norms underlying a discipline’s communicative process will probably be excluded from participation, i.e. will not be regarded as serious contributions and hence not be accepted for publication. This might very well be the case. One should remember, however, that it is not ‘the system’ that is responsible for accepting or rejecting articles, it is people. The review of articles for journal publications are generally carried out by fellow authors. Admittedly, these colleagues might or might not share the concerns brought to the fore in an article. But in the end, this human aspect of scholarly communication implies that causing changes within a discipline is not a matter of overturning a system, but of convincing people. This point indicates why science can be defined as “a self-correcting enterprise which can put any claim in jeopardy though not all at once” (Sellars, cited in Williams, 2001, p. 179). My interpretation of authorship should not lead us to conclude that disciplines may change radically overnight, nor should it lead us to think that authors could just go about writing whatever they want. I agree with Vanderstraeten when he says that strong epistemic forces are working through disciplinary communities (p. 44). Looking from the midst 1960s onwards a convergence has taken place in scholarly communication in education, resulting in an almost monomaniacal obsession with quantifiable outcomes. What I am suggesting, however, is that it is possible to question such a tradition. At one moment one might choose to just ‘add small particles of knowledge’, but at another one might try to bring some basic elements of the tradition up for discussion. My interpretation gives way to the author’s subjectivity in that it offers him a ‘choice’ between different ways of presenting himself to a tradition, i.e. to his audience. 50
The interpretation of authorship developed in this response has consequences for an interpretation of the relationship between different generations of scientific researchers. An author’s choice concerning the kind of author he wants to be entails a responsibility, for the outcome of his choice may affect the course of a tradition2. This implies that ‘new’ generations researchers have a responsibility to reflect on the disciplines tradition, and the role they aim to play in it.
References Brandom, R. B. (2000). Vocabularies of pragmatism: Synthesizing naturalism and historicism. In R. B. Brandom (Ed.), Rorty and his critics (pp. 156-183). Malden, MA: Blackwell Publishing. Foucault, M. (1975). Surveiller et punir. Naissance de la prison. Paris: Gallimard. Foucault, M. (1988). The ethic of care for the self as a practice of freedom, an interview (J. D. Gauthier, Trans.). In J. Bernauer & D. Rasmussen (Eds.), The final Foucault (pp. 1-20). Cambridge, MA: MIT Press. Hemrica, J., & Heyting, G. F. (2004). Tacit notions of childhood. An analysis of discourse about child participation in deciding on regulations in case of parental divorce. Childhood, 11(4), 449-468. Heyting, G. F. (2001). Antifoundationalist foundational research - analysing discourse in children's rights to decide. In G. F. Heyting, D. Lenzen & J. White (Eds.), Methods in philosophy and education (pp. 108-124). London: Routledge. Kuhn, T. S. (1962). The structure of scientific revolutions. Chicago: University of Chicago Press. Lepková, K., Heyting, G. F., & Mulder, E. (2004). De sociale constructie van 'intelligentie'. Reconstructie van het intelligentiebegrip in de Tsjecho-Slowaakse discussie over het pedagogisch gebruik van tests in de periode 1900 - 1939. Pedagogisch Tijdschrift, 29(4), in druk. Williams, M. (2001). Problems of knowledge - a critical introduction to epistemology. Oxford: Oxford University Press.
2
This relates to Foucault’s (1988) idea of ‘the ethic of care for the self’; an idea Foucault developed later on in his career, focusing much more on subjective freedom than he had done before.
51
5 On the shoulders of giants CHRISTOPHER WINCH
The problematic of post-analytic philosophy of education British analytic philosophy of education is, perhaps, best known for its apparently uncompromising attempt to develop a canonical conception of education on the basis of synchronic conceptual analysis.1 The fruits of such work are usually associated with an account of education as the acquisition of intrinsically worthwhile propositional knowledge of appropriate breadth and depth (the early Peters (e.g. Peters, 1966)) and with the development of the rational mind through an acquaintance with the forms of human knowledge (e.g. the early Hirst (circa 1974)). It is also accepted that both views can be combined to form a synthesis, whereby the Peters claim that an educated person possesses knowledge in appropriate breadth and depth can be fleshed out through a Hirstian characterisation of the forms of knowledge.2 Such attempts are widely rejected nowadays, even if the attachment of these authors to a form of liberal education is still endorsed (for an exception however, see Carr, 2003a; 2003b). The central problem for the early Peters-Hirst enterprise is that the mode of derivation of the concept of education has been widely rejected, rather than the central substance of such education, which is still highly influential in the UK. But this leaves the proponents of traditional liberal education with a problem: how to persuade society of the continuing relevance of the liberal educational ideal? One possibility would be to adopt the diachronic or hermeneutic approach, making the justification of education dependent on an account of the historical evolution of the concept of education and its cognates. A problem that immediately occurs to anyone considering this enterprise is 1
In this respect it rejects, like analytical philosophy more generally, the historical or diachronic elucidation of concepts and conceptual networks (see Heyting, 2001). 2 It is important to note, however, that Hirst’s analysis, unlike that of Peters, was intended to be valid for liberal education, not for education as such.
53
that it is evident that there is a plurality of conceptions of education, each with its own conceptual history. We could thus focus on the German tradition of seeing education in terms of Bildung and Ausbildung, which would give us a different conception to that of the English liberal tradition.3 Alternatively, it would be possible to look at the ‘progressive’ tradition instituted by Rousseau and continued through the work of Pestalozzi, Dewey, Neill and others, and come up with a somewhat different interpretation of the liberal tradition. Or it might be possible to look back to the accounts of Plato and Aristotle, and the idea of a ‘gentlemanly’ education to be found in these writers and to provide a hermeneutic account of education on this basis. The last possibility sounds, perhaps, the most attractive one, since it is at least arguable that such an account has a universal appeal, and a present day relevance. The most striking illustration of such a hermeneutic enterprise within the analytical tradition would be the recovery of Aristotelian moral concepts and their re-application to contemporary moral education. But even within this enterprise, conceptual pluralism rears its head. Virtue ethics can be looked at in two ways, first as a moral doctrine about which virtues one should inculcate, second as a meta-ethical doctrine to the effect that character is the central, if not the only moral concept (Hursthouse, 2001). Despite the fact that the two central ancient philosophers of virtue did not explicitly distinguish between these aspects of ethics, it is evident that, within a meta-ethical framework of thinking about morality in terms of the virtues, there is considerable scope for moral variety. We can distinguish, as Nietzsche did, between the morality of the Homeric era and the slave morality first set out by Socrates and then incorporated, with modification, into an ethic of self-hatred and self-abnegation within the Christian tradition. According to MacIntyre, such ideas about morality exist within practices, which are themselves constituted by traditions of thinking and acting, whose telos is given by their internal goods, which are both the aims of those practices and the characteristic ways of achieving those aims (MacIntyre & Dunne, 2002). We can now see a problem that has preoccupied much of the work of Frieda Heyting, namely how one could be able to provide a rational critique of an authentic moral tradition if one doubted its worthwhileness? For might not a Nazi adopt a version of the Nietzschean critique of Christian morality? How would one criticise rather than merely suppress such an enterprise? More generally, all attempts to move away from canonical synchronic analysis in philosophy of education seem doomed to wrestle with the 3
For an historical survey of the Dutch scene, see A. Westerhuis (2006 forthcoming)
54
problem of relativism, which deems rational critique of alternative points of view futile or impossible. To see this problem in sharper focus, it is first necessary to look at the analytical methods of ‘conservative’ philosophy of education.
The conservative analytic project: particularity as generality – early Hirst and Peters Analytical philosophy of education, as it was practised in the 1950s and 1960s, sought to provide a universal conceptual analysis of education and related concepts by an examination of the ways in which those concepts were employed within the English language of the time.4 Peters drew attention to the strong conceptual connections between education, preparation for life, learning and knowledge. He also maintained that education involved a concentration on propositional knowledge, of a certain range of subjects (breadth) and to a certain degree of thoroughness (depth). An educated person enjoyed ‘cognitive perspective’, by which was meant an ability to see the world from beyond the immediate confines of one’s situation and to be able to formulate courses of action that took into account that knowledge. Peters did not distinguish between liberal and other forms of education, although he did distinguish between education and training. For Peters, education turned out to be liberal English education of a certain kind.5 Working independently of Peters, Hirst developed an account of liberal education which, in some ways, put flesh on the bones of the Peters’ account. Although Peters was clear that to be educated is to have one’s rationality developed (this is a large part of what is meant by ‘cognitive perspective’, he did not specify the range of matter to be learned, nor the degree of depth to which it was to be learned, in any detail. Hirst does, however, provide an account of these matters. A rational mind, according to Hirst, is one that is acquainted with all the characteristic types of human knowledge. 4
Peters and Hirst make few references to usage in languages other than English, with the exceptions of Greek and Latin. There is, arguably, an implicitly hermeneutic side to their approach. 5 Some of those who continue to work within this tradition make a distinction between education, which corresponds, roughly to the original Peters conception and schooling, which is a form of preparation, which, although valuable, is not educational (see Barrow, 1981; and Carr, 2003a; 2003b for examples).
55
These are not to be identified with subjects, but underlie them and may, sometimes, coincide with them. They constitute the characteristic ways in which humans have learned to acquire and to classify propositional knowledge. They are: mathematics/logic, natural science, social science, morality, philosophy and artistic knowledge.6 Hirst also gives us an idea of what it is to be acquainted with a form of knowledge. Each of them he tells us, has a core of central concepts, characteristic modes of gaining new knowledge within them and characteristic modes of checking on the truth of propositions (see also Hamm, 1989, ch. 5; Hirst, 1974). Someone acquainted with a form of knowledge will, then, be equipped with at least these features of a form of knowledge and, preferably, some highly specific knowledge within one or more of the forms as well.7 It would, therefore, be possible to describe a London analytic synthesis of what education consisted in, which would be a liberal education in the exercise of one’s rationality, requiring thorough acquaintance with a broad range of subjects to be found, for example, in the curriculum of the grammar or public school.8
Transitions – Peters on democratic education, Hirst on practices. The concept of education and conceptions of education If Peters’ and Hirst’s treatment of education had remained static, then those critics of analytical philosophy of education who complained that it was an ideological justification for a particular élitist and conservative form of education for a gentleman who was, ultimately, a descendant of the Aristotelian citizen, would have been justified. The interesting point about Hirst and Peters, though, is that neither remained satisfied with the ‘London synthesis’, but even more interestingly, the sources of their dissatisfaction, although related, are also different. Take Peters first. Peters’ growing dissatisfaction with his earlier views, which had, perhaps, achieved their fullest expression in Ethics and Education, published in 1966, stemmed both from his realisa6
Early formulations included both religious knowledge and history. The former was dropped, the latter subsumed into social science. Moral knowledge appeared in a later classification. 7 Hirst also thought that reasoning within forms of knowledge was largely form-specific and depended on a good acquaintance with the form. He rejects the idea that context-independent critical thinking could ever be a substitute for such a form specific ability. 8 A notable, and curious, omission is the knowledge of languages, which have traditionally occupied a privileged place in this curriculum, formerly the classical languages, latterly modern ones such as French and German.
56
tion that there was more to education than traditional liberal English education, and also from a growing dissatisfaction with the kind of philosophical analysis that he himself had been practising. Peters came to believe, by the early 1980s, influenced by W.B. Gallie’s article ‘Essentially Contested Concepts’ (1956), that education is an essentially contested concept, meaning that there is no, and cannot be any, canonical, universally valid account . Instead there are various conceptions, more or less incompatible with each other, which give voice to different societal points of view concerning what constitutes a worthwhile education. Peters did not come to believe that the categorical concept of education was completely devoid of content, but he thought that what could be said about it was both limited and underdetermined various possible conceptions of education. Any education involved preparing young people for life, it had to involve learning and the preparation had to be for something regarded as worthwhile. Certain interesting consequences followed from this: first that education was an engagement with values (because of the concern with the worthwhile); second that it had to have aims (since different conceptions of education would express different interpretations of what was worthwhile, and particular conceptions needed to orient themselves towards particular interpretations of the worthwhile); thirdly, education could no longer be seen in liberal terms only, since some preparations for some worthwhile kinds of life would involve preparation for work or employment.9 Hirst’s dissatisfaction with his earlier ideas stemmed partly from a growing realisation that liberal education was not an appropriate preparation for everyone, but also from an increasing emphasis on something that was already implicit in the early work, namely that knowledge is located in human practices (connected with the claim that forms of knowledge involve modes of acquisition and confirmation). Hirst thus came to believe that, although his earlier account was a valid one for a particular conception of education, it was not only not a universal account but was also one that was appropriate only for a relatively small élite of students. Instead, education for most people is an inculcation into rational practices, which were those that adopted intelligible means towards intelligible ends. In MacIntyrean terms, the internal goods of these practices would be intelligible as worth striving for, either because of their intrinsic value, or because they were a means towards practices that constituted intrinsic values. They would also be realisable by means that bore a recognisable and appropriate relationship to the ends of 9
R.S.Peters (1983) Ch. 3: ‘Democratic values and educational aims’.
57
the practice. Someone educated into a practice would be able to rationally reflect on these internal aims and the means that could be adopted to achieve them. They would do more than blindly adopt the routines of the practice without any appreciation of their importance or of what constituted acceptable ways of realising them.10 It would be fair to say that the dominant influences on Hirst became Aristotle (to some extent via MacIntyre) and Wittgenstein. Practices constitute a philosophical given which are not themselves to be subject to critique except in the limited sense that they must be rational if they are to be worthy of adoption by education. Both Peters and Hirst re-oriented their view of philosophical analysis to a modest, but more achievable goal of providing a conceptual framework for thinking about a central human institution such as education. They both thought that it was possible to do this without spelling out in detail what education had conceptually to be. They both became, as a result, far more open to genuine educational possibilities different from those which they had themselves experienced in their own upbringing and which continued to be the major ideological influence on British education. In doing so, they opened the way to a more universalist analytical approach to the subject which also had a proper sense of the limitations concerning what philosophy of education could and could not realistically achieve. British analytical philosophy of education does not, therefore, fully correspond to the view that many still hold of it. But the new approach raises many questions, both about education and the nature of philosophical analysis. Let us take Peters’ later position first. If Peters is right, there are certain universal truths about education, namely that it is a preparation for a life seen by the educators to be a worthwhile life and that it involves learning. But beyond this point, it is far from clear that there is anything universal to be said. Different societies adopt different views of what a worthwhile life consists in and each of these societies may take a different view of what is a worthwhile life for different kinds of individuals within those societies. Conceptual analysis can compare and contrast these different conceptions of education and their relationship with each other. Can it also recommend a particular conception of education? To do so seems to take us beyond analysis and into ideology and politics. However, this does not preclude the use of philosophical techniques to develop arguments for adopting a particular conception of education. The proviso must be, however, that in order not to have 10 Although Hirst does not spell it out, this is what would presumably distinguish between an educated and an uneducated participation in a practice (see Hirst, 1993).
58
the illusion that one is developing a canonical account, one must be aware of what one is doing, and, in order not to mislead one’s audience as to what one is doing, one must signal the point of view from which one is developing one’s position. Similar points can be made about Hirst’s work. If the practices to be recommended are rational, and there is no transcendent content-rich concept of rationality to appeal to in the assessment of the rationality of practices, then it is difficult to maintain that there is a substantial external perspective that we can adopt in our attitude to practices. It seems that their rationality or otherwise must be assessed by some kind of immanent critique, if not absolutely practice-specific, then at least closely related to specific practices or congeries of cognate practices. To do this is to take a position very close to the one suggested in relation to Peters. In order to evaluate practices, one must adopt a practice-specific form of critique which one is obliged to be both open about, and self-conscious of, in order to preserve a proper sense of the limits of what one can achieve through such methods. To sum up, there is a role for the traditional conception of philosophical analysis in the philosophy of education, but it is limited to a search for the limits that any conception of education must conform to.11 But this does not mean that philosophy of education has to limit itself to this austere role, provided that it is open about the commitments that it makes. The view that I wish to defend is that it is possible and rational to adopt this view of the subject. But, in order to defend such a view, it is necessary to deal with a fundamental objection to it and this is where I propose to turn to the work of Frieda Heyting for help in constructing such a defence. The objection is that such a post-analytic analysis and defence of conceptions of education rests on the adoption of an ideological and value specific viewpoint, which one cannot assume that one’s interlocutors will share. And if they do not share one’s assumptions, there is no reason why they should share one’s conclusions, even if they appreciate the force of the arguments put forward in favour of that conception. Isn’t the important thing, after all, to persuade others of one’s assumptions? And, on the conception of philosophy of education being offered here, that is ruled out as one of the tasks of philosophy of education itself.
11 And even here, not all is straightforward; for example, if education is a preparation for life, at what point does one cease to prepare for life and simply ‘live’. The Bildung conception of education, for example, seems to suggest a much more open-ended interpretation of preparation through the course of a human life than does the English conception of liberal education.
59
Radical disagreement and truth: Wittgenstein’s position in On Certainty In On Certainty Wittgenstein (1969) points out that in certain kinds of radical disagreement, dialogue is not possible, because the assumptions necessary for such a dialogue (hinge or riverbed propositions) are not available to at least one of the disputants. The claim is that life is based on unspoken, untaught, but not unlearned assumptions which form the basis for any possible critique within a culture.12 These assumptions range from the universal, such as ‘Humans have bodies’, ‘more than one person exists’ to relatively local or personal ones, such as ‘My name is N’ or ‘I have never been to the Moon’. Not all human cultures share the same set of these unspoken assumptions, although some are shared by all (e.g. the two universal ones above).13 Discussion, argument and dialogue can take place on the basis of such shared assumptions, but otherwise only with difficulty, if at all. It is also possible for two cultures to share universal hinges but not some local and linguistic ones.14 One response to the view of post-analytic philosophy of education developed above is that different conceptions of education rest on different sets of hinge propositions and hence that rational forms of disagreement between different conceptions are difficult to conduct. A Spartan education is conducted on certain, (for Spartans), indefeasible assumptions both about human nature and human values, and a Rousseauian education is based on assumptions that are not only incompatible, but do not even stand in a potentially dialogic relationship with, the Spartan ones. If this is the correct, then postanalytic philosophy of education is a form of rhetoric, based on the attempt to persuade by means other than what are usually called rational arguments 12
The following interpretation of On certainty relies heavily on Rush Rhees’s Wittgenstein’s ‘On certainty’; There like our life, edited with a commentary by D.Z. Phillips (2003, pp. 8889; see Wittgenstein, 1969, § 609-612). 13 Danièle Moyal-Sharrock (2003) offers a fuller typology of hinge propositions. Linguistic hinges include definitions and simple normative statements, personal hinges concern beliefs that constitute one’s personal identity, local hinges are to do with beliefs that allow one to operate successfully in one’s environment and universal hinges are those beliefs that constitute a general human understanding of what it is to be in the world (for detailed examples and references to Wittgenstein’s discussion, see Moyal-Sharrock, 2003, p. 129). 14 Hirst’s account of the structure of inference within forms of knowledge has strong affinities with the account given in Stephen Toulmin’s The uses of argument (1958) which relies on the idea of ultra-localised hinges (inferential warrants) which usually rest on a defeasible empirical basis (backing statements).
60
(i.e. arguments that move from premises, by intelligible and acceptable steps, to conclusions). In such a situation, humour, sarcasm, irony, threats, flattery, the quasi-poetic presentation of utopias, brilliant imagery etc. are the essential tools in the armoury of philosophy of education.15 The adoption of this view of the subject, however, would be tantamount to abandoning a key feature of the analytic and, indeed, the hermeneutic heritage, namely the construction of arguments of the premise-conclusion (pc) form in order to bring disputants round to one’s point of view by the recognition of the quality of the arguments deployed. For many philosophers of education, it would be a counsel of despair.
The postmodern challenge to post-analytic philosophy of education: a radical pragmatist theory of knowledge (involving a pragmatist theory of truth) A related response would be to question whether or not there was any truth in conceptions of education. It is important to note that the position outlined in the section above is not a relativist one. From the point of view of the proponent of a conception of education, there is a correct or even a true view. The problem is that pc-type dialogue cannot always be deployed in its defence because of radical disagreement about certain hinge propositions, the resolving of which cannot be accomplished by the use of such arguments. A relativist view, on the other hand, suggests that there are multiple, equally valid, truths of the matter, or, less drastically, that there is no truth about it, only opinion. This former view will be rejected as it will be argued that alternative conceptions can be considered false, even if they do not stand in a particularly strong dialogic relationship to the one that is true, when for example, they are adopted on pragmatic grounds. The latter position would, for example, be that of logical positivism with regard to value-laden statements such as those putting forward a conception of education. In this case, attempts to resolve disagreement by argument are not just difficult or practically impossible, they are irrelevant. In the former account, which I am going to assume is the more plausible, albeit incorrect, one, one might adopt a 15 Some authors (e.g. Levi, 2000) have argued that this form of argument is much more pervasive than logicians are usually inclined to suppose. For attempts at persuasion through sheer literary brilliance and outstanding analogies one needs to look no further than Plato or Rousseau.
61
pragmatic conception of truth. One adopts as true those propositions that suit one’s purposes.16 And since the purposes of different conceptions of education are different, then it is quite legitimate to adopt a relativist pragmatic conception of truth with respect to those conceptions. The pragmatic conception of truth is not as absurd as it may at first seem, particularly if it is brought to bear at the right level of relevance. Thus, it might be absurd to judge any individual particular empirical proposition to be true on pragmatic grounds, but far less so to adopt a particular framework of judgement on such grounds. It is not unreasonable to suggest that this is, for example, what is done in science when atoms are from one perspective, regarded by analogy with billiard balls, from another, by analogy with miniature solar systems and from another, as nuclei surrounded by fuzzy waves (Gribbin, 1995, ch. 5). On this view of the matter, we can regard discussion within the context of a particular conception of education in terms of coherentist or correspondence-like conceptions of truth, while reserving a pragmatic framework for the conception as a whole. Thus, different conceptions of education may be true, depending from which pragmatic viewpoint they are looked at. From the point of the economy, some form of vocationalism may be true, from that of individual development, some form of liberal education, from a political perspective, some form of civic education. I will, however, reject this suggestion because I do not think that the study of education is so intractable as to warrant the kind of desperate measures adopted by some natural scientists when confronted with, for example, quantum phenomena, and will set out, instead, an alternative to relativism. We do not have to say that different, contrary, conceptions of education are all true, but that, from the point of view of each conception the other is false. But which one or ones are true, appropriate or valid for a society, may be settled outside the educational practices themselves, but within the society in which they occur, possibly according to non-pragmatic criteria.
16
In the more moderate version that I will discuss, the pragmatic conception of truth is applied to those propositions that are constitutive of the conception of education to be defended (the ‘hinges’, if you like), this leaves open the possibility of more conventional forms of truthdetermination within the practices constituted by a particular conception. For the beginnings of a justification of such a view from a Wittgensteinian perspective, see Gerald Vision ‘The truth about Philosophical investigations I paras 134-137’ (2005).
62
Frieda Heyting’s synthesis. Contextual presuppositions of a discourse. Context and purpose relativity In an important paper, ‘Relativism and the Critical Potential of Philosophy of Education’, Frieda Heyting (2004) makes the important suggestion that we can avoid the problems inherent in a potential dilemma between foundationalism and relativism by adopting a contextualised perspective on questions of truth.17 If I understand Professor Heyting correctly, it is possible to develop a non-relativist account of truth which is also non-foundationalist, because the presuppositions on which one’s veridical judgments are made rest on non-universal, but contextually bounded assumptions which are, nevertheless, highly reasonable. Within such a conceptual framework it is possible to make judgements according to correspondence and/or coherence-like criteria. What does this mean for education? There seem to be two important consequences. The first is that within a framework of contextual assumptions (which are, in normal circumstances, pragmatically indefeasible), it is possible to make valid judgements as to truth and falsehood on non-pragmatic grounds. The framework may operate according to the warrants that are licensed by overall contextual assumptions, and these warrants may be wide or narrow ones, according to one’s philosophy of inference. At the limit, they may be constituted by classical logic or, if one prefers, by a more contextualised system of inference.18 Relativism is thus banished from within conceptions of education. Propositions are judged to be true or false according to the criteria developed within a particular contextual framework. For example, within a conception of education that assumes measurable academic achievement to be a major criterion of educational success, it is legitimate to develop value-added measures and to judge the effectiveness of schools on such a basis. For relationships between conceptions of education themselves, the position is more complicated. First of all, it is not impossible, and may be desirable, to assess the validity of a particular conception of 17
It might be thought that coherentism involves a way out of this problem. But if the criterion of the truth of a proposition lies in its coherence with other propositions within the relevant field, and if coherence is to be interpreted as consistency (which seems to be a minimum criterion of coherence), then we seem to be landed with a form of a priori foundationalism in any case, since the criterion for consistency is the Principle of Non-contradiction (see also Van Goor et al., 2004). 18 Hirst’s theory may be developed in either or both these ways. For the first see Mackenzie (1998) Forms of knowledge and forms of discussion, for the second, see Toulmin (1958) for a plausible model.
63
education from outside that conception itself (for example, from a different socio-political viewpoint within that society), or, indeed, independently of any particular conception of education (if a society considers education de novo). If the ‘categorial’ analysis of education is accurate, and education is to be thought of as preparation for life, then conceptions of education can be judged, in terms of their appropriateness as preparations for particular kinds of lives, which may differ according to the view that one takes of individuals or classes of individuals – see the ways in which Plato and Aristotle, for example, differentiated between the different kinds of education that were appropriate for different kinds of people. In other words, the validity and scope of a conception of education is to be judged in terms of the values of the society in which it is a possible preparation for life, and can be deemed as incorrect, or even false according to that society’s conception of the worthwhile. 19 Judgments concerning the validity of particular conceptions of education are, then, for the most part, contextually based. What is more, they are often also purpose-relative. We may thus judge a certain kind of education to be appropriate for civic, but inappropriate for vocational purposes for example, or we may consider it to be appropriate for a member of the ruling élite, but not for a peasant and so on. But these are judgements capable of evaluation from within a society, they are not subjective nor are they even necessarily intersubjective, but objective within the criteria of objectivity that is recognised within that society. Such objectivity may, for example, be obtained through civilisation, rather than society-relative values as a yardstick for subjecting pragmatically acceptable conceptions of education to transcendent critique at times of social crisis (see for example Gray, 1995a; and Taylor, 1989). An example is the abandonment of school selection by IQ testing in Britain in the years after the Second World War. A proposition adopted on essentially pragmatic grounds can be re-evaluated according to different considerations and be rejected according to those latter considerations. Neither does this way of looking at the critique of educational propositions presuppose that judgments must be made from outside the educational conception itself. To the extent that educational practices are rational, they possess criteria according to which educational propositions concerning that practice may be judged. So a critique of education may be immanent, as well as externally generated. Thus a pragmatically constituted truth may be questioned 19
If judged correct/incorrect or valid/invalid, rather than true/false, such a judgment may still emerge as a result of pc argument, according to shared premises and accepted modes of inference, thus being objective.
64
from within the practice according to criteria that now seem more appropriate, for example, because of the growing prestige of research-based evidence. Indeed, it may be somewhat misleading to distinguish to sharply between these different kinds of critique. Social practices do not stand on their own, and they are not intelligible in isolation from the other practices that together constitute a culture and society, which has a certain unity of its own. This is especially true of education, which is a practice with a peculiarly intimate relationship with other practices, as we have seen in considering Peters’ later work. It is thus quite possible that particular conceptions of education themselves may come under scrutiny within societies. It may be that the interpretation of values changes over time, or that certain educational aims no longer seem to correspond to those values, for example. And here again, there are external reference points for judging the validity of such critiques. Relativism is thus banished from a within a society’s conception of education.
Development of this idea: hypergoods and hyperassumptions. Immanent rather than transcendent critique Nevertheless, we might still have reservations. For, after all, are we not merely pushing back the problem of the foundation of our judgements to the assumptions and values on which our societies are built? And, if that is indeed the case, then do we not encounter the problem of relativism at this further remove, not at the level of practices, but at the level of cultures and societies? I don’t think that this is the case for a number of reasons. In developing a way of mounting a ‘transcendent’ critique, without being committed to traditional forms of foundationalism, either empirical or a priori, I shall see myself as developing Heyting’s insight further, thus showing that there are features of education to which every conception must conform. Local and universal hinges - hyperassumptions. The first step is to remind ourselves of the concept of ‘hinge’ or ‘riverbed’ propositions. These are propositions the depth of whose embedding in our way of life, in our practices if you like, is such that they are not formally taught, are rarely if ever, articulated, but are learned as part of the background to our practices. But they are not always immutable; those that Moyal-Sharrock calls linguistic and local hinges may well give up their 65
normative, constitutive role to take up the role of empirical propositions that may be known or doubted, as other assumptions change and as knowledge increases. Thus Wittgenstein’s example that man has never been on the moon was, as he wrote, on the cusp of changing from a local hinge proposition to a false empirical one, due to changes in technology that were well under way at the time he wrote about that example. But other hinges, even other local ones, are apt to stand much more firm, and they constitute ‘warrants’ or localised inference permits that govern the way in which we judge empirical propositions to be true or false.20 Universal and personal hinges however, such as ‘the world has existed for more than ten minutes’ or ‘I am the same person as I was yesterday’ do not change in the same way. But neither are they foundations in the traditional epistemological sense: they are rarely, if ever articulated, it makes no sense to say that one ‘knows’ one and consequently, it makes no sense to say that they are indubitable.21 Critique takes place against the background of hinge propositions that are largely, if not completely, culture and society transcendent. They constitute part of the framework that we employ in our judgements of truth and falsity and, in most cases, to fail to act as if they were true is to forfeit one’s claim to rationality. Civilisation-relative values – hypergoods. In addition to such assumptions, there are also what Charles Taylor (1989) has called ‘hypergoods’, or values that belong to a civilisation or whole series of civilisations. Such values are usually underdetermined for any particular society or culture. It is easy to see that conceptions of justice, truth, rationality and love, for example, differ from each other in different times, places and cultures. But particular values and practices tend to be evaluated against a transcendent, if underdetermined, hypergood. Most of us find it hard to think of what a worthwhile life would be without any element of justice, a regard for truth, respect for rationality and different forms of love playing some role in it. But what these hypergoods actually amount to in any particular society may not only vary from society to society, but also within societies, especially at times of stress and great change. 20
Toulmin’s account of everyday logic used a form of ‘ultralocal’ hinge which did usually depend on an empirical backing statement which, at times, was required to validate the warrant. Some of Hirst’s central propositions within forms of knowledge are possible examples of such ultralocal hinges. 21 For the difference between Quine’s and Wittgenstein’s approach to foundations, see P.M.S. Hacker Wittgenstein and twentieth century analytic philosophy (1996, ch. 7).
66
Limiting concepts Giovanni Batista Vico (1968 (1744)) tried to emphasise the huge variety of human experience and the enormous effort necessary to think ourselves into the assumption of primeval mankind. However, he maintained that, within this prodigious variety, it was possible to discern a stable framework, invariable constants in any human society: reproduction, marriage, birth and death. Vico’s vision, in his New Science of the enormous variety of ways in which humans beings can live their lives, is thus ultimately bounded by biological constraints and characteristic ways of responding to them. Birth, reproduction and death are biological constants in human life and give rise to characteristic human responses, albeit of great variety: marriage, religion and funerary rites being the principal ones with which he is concerned.22 Such limiting concepts form the basis for the understanding of any human society and many of the universal hinge propositions can be identified within the language and culture of societies that are very different from our own. These provide a point d’appui for the more comprehensive understanding of such societies, if interpreted in accordance with the ‘Principle of Charity’ which enjoins us to make interpretations that assume our human subjects are rational unless we have compelling evidence to suppose otherwise, and the ‘Principle of Humanity’, which enjoins us to think of that rationality in human terms, that is, in terms of the kinds of constraints that Vico is concerned with (Grandy, 1973). Hollis has shown, in some detail, how the assumption that other cultures are not constitutively rational, that is, do not operate according to the same logico-linguistic constraints as we do, is barely coherent as an operating assumption. This is not to say, of course, that for example, the role that ‘logic problem stories’ play in their lives, or the ways in which they express fundamental logical distinctions may not be different from ours, but we ought to assume that they are there in their cultures if we are to make sense of those cultures (Bloom, 1975; Hollis, 1995; Levi, 1996).23 Vico’s limiting concepts are of particular relevance to education, since a little reflection will show that, within Vico’s trinity of birth, reproduction and death 22 In the case of ‘primitive ‘ religion, this is interpreted as terror of natural phenomena such as thunder. Vico argues that primeval man’s response of fleeing into caves from such manifestations gave rise to permanent abode and hence to marriage and the family. But see Arthur Schopenhauer The world as will and representation (1985 (1859)) for an alternative universalist account of the significance of religious belief. 23 There is also a vast literature for and against the proposition that illiterate cultures are, in some constitutive as opposed to contextual sense, non-rational. Vico himself seems to have thought in terms of a ‘common sense’ rationality that is common to all humanity (cf. Vico, 1968 (1744), §141-146).
67
there is the universal human practice of bringing up the young and preparing them for life within their society. This is a species constraint, since human young remain immature and comparatively helpless for a long period. Furthermore, the complex of knowledge and skill, together with the complexity of human culture, that the young need to master, require that education, in the categorical sense, be complex and various, even if it is not confined to institutions such as schools and colleges. The enormous complexity and cultural variety that we find in human life should also make it perfectly comprehensible why there should be different conceptions of education corresponding to different views of what a worthwhile life actually is. It should be noted that Vico’s constraints, the principles of charity and humanity and the idea of universal hinge propositions all constitute highly general limits on the possible ways in which we can interpret cultures and forms of thinking that may appear and indeed, in many respects are, very different from our own.24 They are not an injunction to ignore human difference in favour of the imposition of general categories and they should, if anything, make one aware of the possibility of vast variety in the ways in which human beings prepare the young for life, and the kinds of subtle and specialised distinctions in respect of educational concepts that we may expect to find, even in societies that share the same civilisation as our own. I hope that I have now said enough to demonstrate the importance of Heyting’s suggestion that we can contextualise our account of knowledge and critical rationality without falling back on foundationalism in the traditional sense, or on any form of relativism, whether it be subjectivist or intersubjectivist. What now remains is the spelling out of what this vision of contextualised rationality and epistemology means for the future of post-analytic philosophy of education. To do so, I will return to the later work of Peters and Hirst.
24
Some have also argued, I think with a great deal of plausibility, for some basic versions of ‘hypergoods’ to be present within any society (see Gray, 1995b, ch. 6).
68
The future of post-analytic philosophy of education. Connectedness and the particular role of education in making sense of life It has already been suggested that R.S. Peters, at the end of his career, had moved to a view of philosophy of education that made a fundamental distinction between the categorical concept of education, which I have interpreted in a Vichian sense, and culture-specific conceptions of education. Regrettably, Peters never had time to work out this new position in more detail and I think that, apart from the work on educational aims that his later position seems to have inspired, not many philosophers of education have taken up the full implications of what he was working towards. The point that I wish to develop, however, is the possible role of philosophy of education in examining, mounting a critique of, defending and advocating particular conceptions of education. At first sight, it might seem as if Peters’ distinctions pretty well abolish philosophy of education as an analytic discipline with claims to objectivity. On the one hand, one sets out the categorical concept of education and, on the other, one leaves it to educational thinkers, ideologues and policymakers to put flesh on and defend particular conceptions of education. It is true that Plato, Aristotle and Rousseau all thought that they were doing philosophy of education, but were they not really spinning out educational prescriptions based on highly contestable visions of human nature and human society, which they mistakenly took to be of universal significance? I don’t think that this is so, because these writers start by developing a philosophy of mind and an epistemology which although, like any philosophical theory, highly contestable, is nevertheless, put forward and taken to be, of universal significance. But we have become much more aware, since the work of Vico, Marx and numerous of the more enlightened comparative anthropologists, that within general psychological, logical and epistemological constraints, an immense variety of rational institutions are possible. A modern Plato or Rousseau would be well advised to take account of these constraints. But it is possible to do so; it requires a self-reflexive awareness that what one is doing is itself culture-bound, although it may have more general implications. It is also likely to be, to a considerable degree, imbued with ideological assumptions with obvious links to important values within one’s own culture. What it need not do, however, is to lack objectivity. To be objective is not to adopt a ‘view from nowhere’ as if this were possible, but to submit to cri69
teria of judgement that are widely recognised as applicable, if not in all cultures necessarily, then at least in large stretches of our own civilisation, and, if the arguments of Hollis and others are correct, well beyond. We may expect from philosophers of education, then, even when they are advocating or defending a particular conception: rigour, attention to the views of others, consistency and quality of argument. We should also expect from them a considerable degree of self-reflexiveness, that is; a recognition that the assumptions that they base their arguments on are themselves often relatively local and not necessarily accepted by everyone. In this sense, their work will have a conditionality that, perhaps, was not a feature of earlier philosophy of education. But it will be none the worse for that. So it seems to me that the programme that R.S. Peters sketched out at the end of his career has a promising future if the challenge be taken up. The legacy of Paul Hirst’s later work is a more complex matter. I do think, however, that just as the earlier Hirst and Peters formed a synthesis the ‘London view’, so the later views of both form the basis of an updated synthesis. But in order to spell this out, more work needs to be done on the Hirstian concept of a practice. Let me begin by saying that one of the major virtues of Hirst’s concept of a practice is its relative openness. For Hirst, practices are relatively permanent collective human endeavours. Rational practices are those that involve intelligible ends and intelligible means of reaching those ends. The Hirstian idea of a practice, like his forms of knowledge, is essentially hermeneutic, it is an interpretative tool rather than something with a firm instantiation in reality. Hirst does not get involved, like MacIntyre, in discussions of what is and what is not a practice. In this respect, his concept is more akin to the Wittgensteinian device of a language game, or a form of life, rather than something reified like a tradition or an occupation. However, even this hermeneutic concept of a practice, like language games and forms of life, leaves out something important, that Vico and Marx realised and which Wittgenstein’s student Rush Rhees (1970) tried to develop in his important paper ‘Wittgenstein’s Builders’. The point can be simply stated: human life has a unity that extends beyond and connects different practices, however we conceptualise these. If we want to understand this unity, we should think of Vico’s limiting concepts, of Hume’s conditions of moderate25 scarcity, which, he argues, give rise to our concept of 25
Vico, like Marx (and perhaps inspiring Marx), thought that the reproduction of human life and of cultures involved class struggle, the ancient traces of which remain in contemporary languages. Unlike Marx, he did not try to develop a schematic, rather mechanical connection
70
justice, or to the Marxist idea that humans do not just produce their means of existence (which is an important point), but the conditions of that existence, which may extend beyond strictly economic considerations. In other words, it can never be enough to think of education in terms of inculcation into particular practices, but one needs to consider it as an inculcation into a human form of life, with a certain unity of culture and outlook but, nevertheless, with certain irreducible internal or conceptual connections between its different aspects. This unity, together with the fact that human discourse within one practice (Rhees (1998) pays particular attention to ‘conversations’, as they particularly exemplify this point) has a bearing on what is said elsewhere in different practices, must be appreciated in order to properly understand how a society thinks, feels and operates . Indeed, it is a condition of understanding the society one lives in, and of one’s own life and that of other people in the society making sense, that one appreciates the way in which conversations in seemingly disparate areas of life have a bearing on one another.26 This is particularly striking in the case of the categorical concept of education. As was noted, all human societies bear and bring up their young. In doing so, they attempt, not only to bring them up alive and healthy, but also so as to secure the conditions of the reproduction of that society in a form recognisable to them. They attempt to do this, not only in economic terms in the means of subsistence (although this is very important and should not be overlooked), but also in terms of culture, religion, morality, marriage, politics, art, science and value.27 Education in the broadest sense is absolutely central to this, as well as to the equally important role of developing each human being as a distinct individual.28 Furthermore, it is not possible to understand the institutions of education in any particular society without understanding the ways in which they link to the practices for participation in which they are a preparation. There is thus a conceptual connection between between the different spheres of human life in order to produce a highly particular vision of the causal relationships between the different elements of society. More helpful in this respect is the economist Friedrich List, who argued that the reproduction of society was secured by the productive powers of that society, which included morality, religion and other institutions, as well as more specific and specialised economic ones (see List, 1991 (1842)). 26 See G. F. Heyting, B. Kruithof en E. Mulder ‘Education and Social Integration: On Basic Consensus and the Cohesion of Society’ Educational Theory, 52, 4, pp.381-396, for a ‘spontaneist’ account of this based on Mandeville. 27 In the case of some societies, this may involve magic, and not science, for example. 28 See Dietrich Benner’s Wilhelm von Humboldt’s Bildungstheorie (2003) on this point about the dialectic of social participation and individuality.
71
education and civic life, education and religion and education and the economy (to name just three) which it is the task of the philosopher of education to develop, both in terms of the categorical concept but also in terms of particular, culture and value-specific, conceptions of education. Education, then, is complex, pervasive and intimately connected, conceptually as well as factually with most, if not all, aspects of life. The missing piece in a Hirstian concept of education in its broadest sense, as an inculcation into human practices, is a clear account of how practices link together in intelligible human life, and the role that education plays in this linkage. These reflections bring me to my final point. The study of education in general, and sadly, too often, the study of philosophy of education in particular, are often thought of as insular and self-absorbed, also showing a lamentable ignorance of the connection between education and other social institutions. I hope that I have shown, drawing on Frieda Heyting’s work and using her non-insularity as an example, that it is absolutely necessary that this should not be the case. Post-analytic philosophy of education could have the most exciting future as a culturally sensitive, ideologically self-reflexive but objective study of one of the central features of any form of human collective life (cf. Hamlyn, 1985).
References Barrow, R. (1981). The philosophy of schooling. Brighton: Harvester. Benner, D. (2003). Wilhelm von Humboldt's bildungstheorie. Weinheim and Munich: Juventa. Bloom, A. (1975). The linguistic shaping of thought. New Jersey: Erlbaum. Carr, D. (2003a). Making sense of education. London: Routledge. Carr, D. (2003b). Philosophy and the meaning of 'education'. Theory and research in education, 1(2), 195-212. Gallie, W. B. (1956). Essentially contested concepts. In Proceedings of the aristotelian society (Vol. 1955-1956). Grandy, R. (1973). Reference, meaning and belief. Journal of Philosophy, 70, 439452. Gray, J. (1995a). 'Agonistic liberalism' in 'Enlightenment's wake'. London: Routledge. Gray, J. (1995b). Enlightenment's wake. London: Routledge. Gribbin, J. (1995). Schrödinger's kittens. London: Orion. Hacker, P. M. S. (1996). Wittgenstein and twentieth century analytic philosophy. Oxford: Blackwell.
72
Hamlyn, D. W. (1985). Need philosophy of education be so dreary? Journal of Philosophy of Education, 19(2), 159-. Hamm, C. (1989). Philosophical issues in education. Brighton: Falmer. Heyting, G. F. (2004). Relativism and the critical potential of philosophy of education. Journal of Philosophy of Education, 38(3), 493-510. Heyting, G. F. (2001). Introduction. In G. F. Heyting, D. Lenzen & J. White (Eds.), Methods in philosophy of education (pp. 1-12). London: Routledge. Hirst, P. H. (1974). Knowledge and the curriculum. London: Routledge. Hirst, P. H. (1993). Education, knowledge and practices. In R. Barrow & J. White (Eds.), Beyond liberal education. London: Routledge. Hollis, M. (1995). The philosophy of social science. Cambridge: Cambridge University Press. Hursthouse, R. (2001). On virtue ethics. Oxford: Oxford University Press. Levi, D. (1996). Why do illiterates do so badly in logic. Philosophical Inverstigations, 19(1), 34-54. Levi, D. (2000). In defence of informal logic. Dordrecht: Kluwer Academic Publishers. List, F. (1991 (1842)). National system of political economy. New Jersey: Augustus Kelly. MacIntyre, A. C., & Dunne, J. (2002). Alisdair MacIntyre on education: In dialogue with Joseph Dunne. Journal of Philosophy of Education, 36(1), 1-19. Mackenzie, J. (1998). Forms of knowledge and forms of discussion. Educational Philosophy and Theory, 30(1), 27-50. Moyal-Sharrock, D. (2003). Logic in action: Wittgenstein's logical pragmatism and the impotence of scepticism. Philosophical Investigations, 26(2), 125-148. Peters, R. S. (1966). Ethics and education. London: Routledge. Peters, R. S. (1983). Essays on educators. London: Routledge. Rhees, R. (1970). Discussions of Wittgenstein. London: Routledge. Rhees, R. (1998). Wittgenstein and the possibility of discourse. Edited by D. Z. Phillips. London: Routledge. Rhees, R. (2003). Wittgenstein's 'On certainty'; There like our life. Oxford: Blackwell. Schopenhauer, A. (1985 (1859)). The world as will and representation (Vol. 1, Bk. 4). Mineola NY: Dover. Taylor, C. (1989). Sources of the self. Cambridge: Cambridge University Press. Toulmin, S. (1958). The uses of argument. Cambridge: Cambridge University Press. Van Goor, R., Heyting, G. F., & Vreeke, G.-J. (2004). Beyond foundations - signs of a new normativity in philosophy of education. Educational Theory, 54(2), 173192. Vico, G. B. (1968 (1744)). The new science (T. G. Bergin & M. H. Fisch, Trans.). Ithaca: Cornell University Press. Vision, G. (2005). The truth about 'Philosophical investigations I' paras 134-137. Philosophical Investigations, 28(2), 159-176.
73
Westerhuis, A. (2006 forthcoming). The role of the state in vocational education: A political analysis of the history of vocational education in the Netherlands. In L. Clarke & C. Winch (Eds.), Vocational education: International perspectives and developments. London: Routledge. Wittgenstein, L. (1969). On certainty. Oxford: Blackwell.
74
6 On the grounds of arguments. Response to Christopher Winch LEONIE LE SAGE
Christopher Winch discusses a central philosophical-educational question: ‘is rational criticism of educational practices possible?’. This question is also frequently addressed in the work of Frieda Heyting (e.g. 2004). The problematic question Winch thinks philosophy of education especially needs to answer is, in his own terms: how can one “criticise rather than merely suppress” (p. 54) a different moral educational outlook? He relates this concern to the writings of two ‘giants’ in the philosophy of education: Richard Peters and Paul H. Hirst. Their projects, Winch argues, can be seen as attempts to unite the thesis that there is truth in education with the thesis that educational practices are heterogeneous. Education is in both views, according to Winch, always an introduction into the rational aspects of a practice or a preparation for a worthwhile life within that practice, while what is considered to be rational or worthwhile (and hence to be the aim of education) may differ over various practices. Winch’s problem with these views is that since a practice constitutes its own standards of rationality, criticism on grounds of different (external) rational standards is not possible. Legitimate criticism can only be criticism from within, not from the outside, for there are no such external rational standards. And this leads to the (undesirable) relativistic position that philosophy of education cannot possibly rationally scrutinize particular conceptions of education. Considering the conference theme ‘Grey wisdom: compliance or opposition between generations?’, the relationship with Winch’s thesis is clear: the unintended heritage of predecessors such as Peters and Hirst is the problem of relativism. Winch’s solution to this problem draws partly upon Heyting’s ‘context-dependency’ view on rational argument. According to Heyting (2004), rational criticism is possible because we can articulate and therefore criticise the presuppositions that underlie our truth- and value judgements. Although our truth- and value judgements are dependent on context, we can 75
trace the framing presuppositions of this context and question their legitimacy. A critical philosophy of education, then, will involve “analysing and making explicit criteria and conceptions that are implicitly presupposed in specific contexts, thus making them accessible for discussion – outside philosophy” (p. 506, italics in original text). Winch wants to go a step further because one could still maintain that this leads to relativism “at the level of cultures and societies” (p. 65). For we can criticise parts of our own culture, but can we criticise practices that do not share the same presuppositions? What makes rational (transcultural) criticism possible, according to Winch, is the fact that we can assume that people in other cultures are rational (the Principle of Charity, p. 67) and that they experience the same human ‘constraints’ (the Principle of Humanity, p. 67), which (for instance) can be psychological, logical or epistemological in nature (p. 69). An example of such a constraint is the fact that humans are born without the skills needed to survive (a biological constraint). The variety of rational and worthwhile educational practices is limited by these constraints, they constitute “a unity that extends beyond and connects different practices, however we conceptualise these” (p.70). Rational criticism is thus not only possible from within, but also from the standpoint of the human constraints: given these constraints a large amount of variety is possible, but some practices are not rationally justifiable. At this point, I first want to express my appreciation of Winch’s highly informative paper on this subject and my respect for the distinctiveness of his view. However, I would like to point out a problem with his position: it paves the way for objective, but at the same time trivial criticism. For substantial criticism one needs to refer to one’s own conception of a worthwhile life, and as a result, the problem of relativism reappears through the back door. Let me explain. The human constraints that form the boundary of what can be a rational educational practice are introduced as objective constraints. A biological constraint like the earlier mentioned lack of skills of the human infant is an example of such a constraint. There seem to be a lot more of these biological or physical constraints: for instance, that we cannot be in two places at the same time, we cannot visit the sun, we cannot live for three hundred years, et cetera. As these examples make clear, to dwell on biological or physical constraints is a bit tedious; these limits are never really questioned, as they are obvious. But if these constraints really were that obvious, what would be the need for criteria for rational criticism based on these constraints? They are obvious, perhaps indisputable; hence we will not disagree on the de76
mands these claims set in the first place. Discussions about the demands of these claims can therefore only lead to trivial conclusions. Moreover, these types of constraints probably leave too much room for a variety of practices we perhaps think we have reason to criticise. For instance, I do not really know if these constraints constitute criteria on the basis of which rational criticism of the practice of Nazi Germany is possible. Concerning the possibility of and the need for rational criticism then, it would probably be more interesting to look at the less ‘naturally given’ constraints like (notably) the psychological constraints Winch mentions. Now how are we to understand these psychological constraints? Winch states that “within general psychological (…) constraints, an immense variety of rational institutions is possible” (p. 69). However, he does not elaborate on what is and is not possible within a psychological constraint. The most plausible interpretation I can come up with of how to understand these psychological constraints is that they set limits to a psychologically healthy life. When we think that a life without contact with other human beings is perhaps biologically possible, but not psychologically possible, because the person would go mad, we refer to a psychological constraint. Rational criticism, then, is possible if we can show that a particular practice lacks the conditions persons need to lead a psychologically healthy life. But what is a psychologically ‘healthy’ and ‘unhealthy’ way of life? Are judgements about what is and is not a psychologically healthy or sane life really objective? What is considered to be sane and insane or psychologically healthy and unhealthy seems to differ over cultures, and these different views seem to be constituted by different conceptions of what a worthwhile life is. I may believe that a particular space of autonomous choice is necessary for a psychologically healthy life, precisely because that makes it possible to lead a worthwhile life. In other cultures however, persons (e.g. women) lack this space of choice. The possibility of free choice is in these cultures not regarded as necessary for a psychologically healthy life. On the contrary, one might advocate a lack of choice because that guarantees that, say, women indeed lead a worthwhile life (for instance: they do not fall into sin). What is considered to be a psychological constraint therefore depends on one’s conception of a worthwhile life. But what a worthwhile life is forms part of the question that we want to answer! Winch’s argument is that we can rationally criticise (that is: legitimately question the worthwhileness of a practice) on the basis of psychological constraints. But if psychological constraints are constituted by conceptions of a worthwhile life, then these constraints cannot function as objective grounds to criticise a divergent con77
ception of a worthwhile life. Any attempt to base criticism on psychological constraints is bound to fail, because different views on what a worthwhile life is constitute different views on what a psychological constraint is. Consequently, there is relativism on the level of psychological constraints as well. Winch perhaps is aware of this problematic aspect of his position, for in his elaboration on the objectivity of philosophy, he states that it is constituted by “criteria of judgement that are widely recognised as applicable, if not in all cultures necessarily, then at least in large stretches of our own civilisation” (p. 70). This, however, seems to be a rather different view on the objectivity of rational criticism from the view that rational criticism is possible because there are objective constraints that determine the limits of what exactly can count as a worthwhile life. Rational criticism, according to this view, is possible because there is a large agreement about what the criteria of rational criticism are. But what if there is no such agreement? Can we criticise instead of merely suppress a practice whose basic assumptions about what a worthwhile life is differ from ours? It seems to me that on Winch’s account, we cannot. Although agreement on the criteria for determining whether or not something is worthwhile or rational can indicate that these are the correct criteria, agreement by itself is no guarantee of correctness. Hence, Winch’s position cannot rule out suppression and relativism at the same time. It is not in my power here to articulate a serious alternative or supplement to Winch’s view. I do think that Winch’s view on rational criticism and philosophy of education points us in the direction of a solution to the problem relativism poses for the critical function of philosophy of education. He, as well as Heyting, argues that philosophy of education is primary reflection on the assumptions of actual educational practices. I would like to stipulate that the philosopher him- or herself is engaged in a particular practice: that of reason giving. To be engaged in philosophy is to make a commitment to reason, argument, logic et cetera, and not for instance to aesthetics, religion, tradition et cetera. This implies that philosophers in particular should be open-minded and amenable to counter-arguments. The problematic question that opened this enquiry is how one can criticise instead of merely suppress a practice. The answer is that suppression is not implied by criticism - on the contrary - if one is committed to a quest for the best reasons. Spiecker (1991) and De Ruyter (in press) argue that liberal education (teaching the child to take the rights of others seriously, to be open-minded towards different moral outlooks) is the opposite of indoctrination (teaching the child that 78
there is only one truth, to refuse to take other outlooks seriously). The same can be argued with respect to being engaged in the practice of seeking and giving reasons: by definition this is the opposite of suppression. What this demands of interlocutors is that everyone is equally concerned with coming up with the best reasons. This is not a knockdown argument against relativism, but in arguing that philosophy, as long as it is an open-minded quest for the best reasons, is the opposite of suppression, relativism looses its sting. I am fully aware that this suggestion on how to interpret philosophy and the challenges of relativism raises at least a lot of questions. Nevertheless I would like to end not with a further elaboration or clarification, but with a quite different matter related to the conference theme. In the earlier quoted passage by Heyting about the task of philosophy of education to make accessible implicit criteria, she italicizes the terms ‘outside philosophy’. Winch, in his final remarks, states that philosophy of education is often thought of as (amongst other things) “self-absorbed” (p. 72). I think that the concern that speaks from these remarks is that philosophy of education has perhaps neglected one of its main tasks: to be involved in the debate about educational practices. The theme of this conference is how we should relate to our philosophical ancestors. The question for contemporary philosophy of education however, does seem to be above all how we should relate to other specialists in the field of education or to society in general in its concern for educational practice. Philosophy of education in the last few decades seems to have been focusing strongly upon philosophy and less on education. It seems to me that we should oppose this ‘grey tradition’ and be engaged in the process of argumentative reflection on education and educational practices again.
References De Ruyter, D. (in press). Whose utopia/which ideals? In M.A. Peters, & J. FreemanMoir (Eds.), Edutopias: Utopian thinking in education. Rotterdam: Sense Publishers. Heyting, F. (2004). Relativism and the critical potential of philosophy of education. Journal of philosophy of education, 38(3), 493-510. Spiecker, B. (1991). Indoctrination: the suppression of critical dispositions. In B. Spiecker, & R. Straughan (Eds.). Freedom and indoctrination in education. London: Cassell.
79
7 Education, history, and human nature JAN BRANSEN
There is an obviously correct and trivial answer to the main question of this conference: the new generation should of course conform to the old generation if the old are right, just as they should of course oppose the old generation if the old are wrong. This answer need not disappoint a contextualist such like Frieda Heyting, because it raises a whole array of important questions concerning major epistemological and normative themes, such as how we can decide whether the old are right or wrong without buying into a generational bias. In this paper I shall not defend the trivial answer nor pursue the route it opens up. Instead, what I intend to do in this paper is to try to understand the question. Why would we have an interest in asking, and in answering, this question? And I shall argue that we have to make sense of the relation between education and history if we want to understand the question that is the central theme of this conference. My focus shall be broad, and general, and very abstract. I shall discuss ‘education’ and ‘history’ as modes of being human, as symbolic forms, to use a phrase from Ernst Cassirer (1923-29, 1944), or as forms of life, as Wittgenstein puts it (1953). My interest is not in education as a specific social institution, nor in history as a specific academic subject, or as mankind’s process through time.1 I shall rather use the terms ‘education’ and ‘history’ to denote ‘anthropological constants’, a term that I shall use to refer to defining features of being human. There are three different kinds of relations between education and history of which I shall discuss those two that are least familiar in educational departments. The relation that is familiar is the one in which education is thought of as a product of history, and thus as having a history of itself that is the topic of historians of education. The other two relations that I shall explore are: education as a response to history, and education as producing history. 1
As the term is used in Hegelian circles, cf. Fukuyama, 1992.
81
My claim shall be that once we understand these two kinds of relations between education and history, we shall be able to understand why we (early 21st century philosophers of education) should like to know whether the young should conform to or oppose the old, but also why the question is ill conceived. I shall first, however, set the stage for this argument.
History as an anthropological constant A deep distinction between nature and history runs through modern culture (Margolis 1993). The assumption is that things either have a nature or a history, that the properties of a thing are either determined by the thing’s nature or its history, and that understanding the thing’s behaviour is either a matter of grasping its nature or its history. The distinction runs through the sciences as well, unsurprisingly, dividing the academy into the Naturwissenschaften and the Geisteswissenschaften, leaving the social and behavioural sciences behind with a very uncomfortable embarrassment, due to the fact that it is our human nature to have a history, to say it paradoxically. This idea of history as an anthropological constant, a defining feature of what makes us human, can be understood as an ontological, epistemological or psychological claim. The idea could be that it really is the case, ontologically speaking, that human beings have a history rather than a nature, as might be inferred from Sartre’s famous claim that our existence precedes our essence (Sartre, 1946). But the idea could also be taken in a less demanding way, implying merely that it is epistemologically more reliable, or perhaps even merely that it is psychologically more economical, to think of human behaviour as determined by the individual’s history rather than the species’ nature. I intend to remain silent in this paper on how to interpret the idea of history as an anthropological constant. That is, I shall assume the relevance of a striking contrast between exemplars of homo sapiens and exemplars of another nature. I shall assume the contrast to imply a difference between how best to make sense of similarities and differences in the inner and outer make-up of exemplars of either kind, a difference related to the claim that all creatures except humans are best conceived of as having a nature in virtue of which each exemplar of a kind runs through cycles of existence that have a determinate, stable, invariant structure, such that each new exemplar displays a strikingly similar inner and outer make-up. In contrast, exemplars of 82
human nature are best conceived of as running through history, in virtue of which each new exemplar displays an inner and outer make-up that is determined by the specific trajectory of this individual human being through its environment over time. This is the rough idea that stands in need of considerable refinement that I shall merely indicate now, and shall come back to later in this paper. A first issue that needs a more precise discussion is the issue of time scales. That is, we might need a different time scale in the case of human nature to discover its determinate, stable and invariant structure. Determinacy, stability and invariance are, after all, temporal properties. They always presuppose a temporal index, and therefore the invariance of the nature of any creature, say that of butterflies, is an invariance merely within a certain time-scale. This might be understood to mean that the distinction between things with a nature and things with a history is a mere gradual distinction. That is, on a larger timescale, say a couple of billion years, creatures with a nature, such as butterflies, will have a history. And, the other way around, perhaps human beings will merely have a nature if we take the time scale to be a couple of days. A second issue that needs more discussion is complexity. It might be that the variation in inner and outer make-up is not so much a matter of a human being’s history as well as a matter of the complexity of human nature. We can imagine that due to sheer complexity it would be better to investigate human nature on a sub-personal level, cutting down each exemplar of human nature in large numbers of parts - organisms, perhaps, or even smaller. To get the idea: think of how difficult (even absurd) it would be to investigate the entire continent of Africa as if it were one exemplar of a particular nature, displaying an inner and outer make-up that we would assume to be the result of a determinate, stable and invariant structure. In such a case it would of course make epistemological sense to break Africa up into millions of independent parts, each of which might be an exemplar of a particular nature. One might think that this would entail an important loss of insight into the nature of Africa as a whole, but thinking of this insight as intelligible at all might be a mere chimera. And the same might hold for human nature: too complex to make sense of at the personal level, without this having much to do with human historicity. A third issue that would need more discussion is externalism (Wilson, 2004), both diachronic and synchronic. It might be that the variation in inner and outer make-up of human beings is not so much a matter of their history (simply understood), or their nature, but primarily a matter of how the individual is related to its current environment as well as to its past and future. 83
The idea is here that there might be intrinsically relational properties (such as being a Dutch philosopher, a medieval monk, a guitarist, a black person, a woman, or a child) that cannot be reduced to well-determined individual properties, as a consequence of which we cannot determine the inner and outer make-up of any single exemplar of human nature if we fail to take into account a wealth of information about this individual’s environment, past and future. Including an individual’s past and future, i.e. including diachronic externalism, is of course a way to include human nature’s historicity. But in mentioning the three issues that require more attention to understand more fully what it means to say that history is an anthropological constant, I do not wish to deny nor assert that historicity is an essential feature of, and only of, human nature. I simply wish to outline the issues that will be involved in any attempt at understanding the meaning of the claim that human nature has a history. And I have an interest in this claim merely because on this occasion I want to understand why an interest in education raises the question whether the younger generation should conform to or oppose the older generation.
Education as an anthropological constant To complete the stage setting, let me also say a bit about education as an anthropological constant. The idea embraces a couple of related features: (1) that human beings live their lives in situations involving more than one generation; (2) that power, authority and expertise is unevenly distributed among these different generations; (3) that human beings have rational capacities that allow them not merely to respond appropriately to environmental requirements, but also allow them to care for the appropriateness of their own responses (Pettit, 1993); and (4) that as a consequence the relations between the generations display a lot of “downstream epistemic and moral engineering”.2 Downstream epistemic and moral engineering is basically what teaching is. It is what makes education differ from merely a learning environment. It consists in the construction and restructuring of environments by the adult generation such that the younger generation can acquire
2
Cf. Sterelny 2003, who discusses ‘downstream epistemic engineering’ as an issue in the philosophy of evolutionary biology.
84
and develop relevant epistemic and moral capacities with more speed and success. Downstream epistemic and moral engineering makes sense precisely because human beings have rational potential. This is not the place to substantially discuss this feature of human nature. Just a word will have to do. The assumption here is two-fold (Pettit, 1993). Firstly, human beings are intentional agents that can act on their desires and beliefs. They can represent their environment internally (which is what their beliefs do), and thereby find more appropriate means to reach their goals (assumed to be present in their desires). On top of this intentionality, which human beings probably share with some of the higher animals, comes a reflexive level, that I have called man’s educatability (Bransen, 2003). It consists in a concern for the (truth-tracking) quality of one’s beliefs and the (good-tracking) quality of one’s desires, in virtue of which human beings can revise their own beliefs and desires so as to improve their own attuning to their environment. These are sophisticated capacities that require a lot of time to develop, which explains why there is a salient uneven distribution of these capacities (and relatedly an uneven distribution of power and authority) among the different generations. This asymmetry gives the older generation the lead to support the young in developing capacities that are crucial to their living a successful human life. Downstream epistemic and moral engineering need not be explicitly intentional. A lot of research has shown, for instance, that the way adults speak to babies, and almost can’t help speaking to babies, involves a lot of completely unconscious engineering that allows babies to pick up the language much more easily (Gopnik et al., 1999, p 128-132). But besides these varieties of what we may call natural nurturing there is of course a lot of downstream epistemic and moral engineering that is intentional, and that is the topic of, and often also the result of, all the evidence-based intervention research that goes on in our contemporary departments of education. When I speak of education as an anthropological constant, I do not intend to restrict what I shall say to intentional downstream epistemic and moral engineering, but I shall be speaking of asymmetrical educational arrangements involving more than one generation. And it is the necessary presence of these different generations in education that raises the need to introduce a discussion of history.
85
Education as a response to history Suppose it is true that to understand a human being’s behaviour we should be acquainted with the individual’s history, rather than with the individual’s nature. Suppose, that is, that the appropriate time-scale for human behaviour, the complexity of human nature and the diachronic and synchronic externalism of human life is such that one can only make sense of the inner and outer make-up of an individual in terms of the individual’s trajectory through its environment over time. What would that mean for each of us in our daily encounters with other people? Of course, we are acquianted with the history, or at least with long stretches of it, of quite a number of people we meet regularly. Collaborating with these acquaintances would be possible, and probably not much of a problem. But what about the strangers we meet too, day after day? If the assumption carries substantial weight, the implication should be that we wouldn’t have the faintest idea about what to expect and how to respond to such a stranger’s behaviour. Running into an unfamiliar animal or a strange artefact need not be much of a problem to us, given that we can derive all kinds of predictions from assumptions about these things’ natures. But if we cannot rely on such assumptions about the stranger’s nature, because he doesn’t have a nature but only has a history we fail to know, running into a stranger could be extremely disturbing. This conclusion is of course based on the assumption that each individual’s history differs substantially from those of other people. This seems to be a justified assumption as long as we entertain the thought that there is an important difference between things with a nature and things with a history. After all, if individual histories don’t make much of a difference to the individual’s inner and outer make-up, the distinction between things with a nature and things with a history would be much of a distinction without a difference. But the assumption introduces an interesting option too: would it be possible for human beings to have a very similar history? This allows for some rather speculative observations. Perhaps, indeed, most of the exemplars of human nature that have lived their lives on earth so far did go through considerably similar trajectories, due to very stable and invariant material and cultural environments. Perhaps it is only in modern times, with all its emphasis on individuality and globalization, that people start to have significantly different individual histories, as a consequence of which it is for the first time in mankind’s history that the fact that we have a history rather than a nature makes a difference to our prospects for living a successful life among strangers. 86
Two further options seem available along this rather speculative line of reasoning. The first option is to link up with developmental psychology, and to explore ways to rethink individual human histories in terms of fixed developmental pathways determined by human nature. If developmental psychology could come up with a comprehensive theory of how specific developmental trajectories belong intrinsically to the way human nature manifests itself in each new exemplar, we would not need to be upset by modern culture’s support for individuality, globalisation and historicity. We could be happy to let a thousand flowers bloom, and still be confident that we would be able to get along with strangers whose histories are unknown to us. A first encounter should suffice to give us a reliable indication of the stranger’s developmental stage and our knowledge of developmental psychology could do the rest to guarantee successful cooperation. Of course, this is nothing but an extremely overoptimistic dream, that is definitely not part and parcel of developmental psychology’s explicit self-image, but it seems to me an important though largely implicit regulative ideal of developmental psychology as a real and promising natural science. The second option is to trust in what we might think of as a typical modern counterpart of human historicity: education. The speculative suggestion here is that the discovery of our historicity should have alarmed enlightened, socially committed intellectuals to set up education as a most appropriate response. Such a response is based on the acknowledgment that newborn exemplars of human nature are the most disturbing as well as the most vulnerable strangers, and both at once because of their lack of history. Because they don’t have a history, the behaviour of babies is unintelligible, both unpredictable and confused. And, because they don’t have a history, they need to cope with life’s surprises without the support of their own rational capabilities. The contrast between newborn human beings without an individual history and those of the older generation should have created in the older generation an awareness of their specific educational responsibility, and of the feasible gains for mankind as a whole of an educational programme that would support the younger generation to acquire a history that would facilitate their well-being and well-functioning. Thinking of education as a proper response to man’s historicity is a wellknown theme of the Enlightenment. In elaborating further on this theme I would like to discuss two issues. The first is the intentional, as well as rational, character of downward epistemic and moral engineering, in virtue of which we can think of environments, and of the individual trajectories through these environments, as accessible for design. The idea here is that 87
recognizing that human beings have a history rather than a nature entails a dramatic emancipatory momentum, because if we all have a history, rather than a nature, our inner and outer make-up is not simply fixed and given, but is the result of a process that is in principle accessible to the manipulating efforts of intentional engineering. Whereas we might have lived our lives in pre-modern times in stable and invariant environments, without any awareness of our capacity for rational control (in “self-incurred immaturity” as Kant (1784) maintained), the discovery of history is not merely the opening up of diversity, confusion and unpredictability, but does rather allow for the recreation, through rational control, of new, stable, designed, educational environments. Education allows us to receive our historicity as exceptionally good news, because it entails the promise of us all intentionally having the same history, a history that will be for the better, that will support each newborn human being to develop the capacities needed to live a good human life. Connected to this is a second theme that is, apparently, crucial to education as a response to history - at least in its enlightened conception - but that also, or so I shall argue, fails to appreciate the depth of human historicity. The theme is maturation, entailed in the idea that education is a process executed by the older generation and experienced by the younger one; a process that can and will be completed once the newborn child becomes mature, i.e. once the younger generation becomes the older one. One can interpret the idea as if educators accept that human beings, at least the younger ones, have a history rather than a nature, but that there is also a post-historical phase in each human life, the phase of adulthood. The idea then becomes that it is only during childhood that man’s history is critical in determining his inner and outer make-up. That is, the idea seems to be that adults have a history, in the sense of a past, that can be known as an acquired, second nature (to use a phrase made popular by McDowell, 1994), stored and available in his attitudinal and behavioural habits, virtues and principles of rationality. Thinking of childhood as an educational process that terminates in the second nature of adults is a very powerful way of incorporating the elusive plasticity of man’s historicity in an optimistic conception of human history as progress. But it is a conception that presupposes controversial assumptions about the relevant time-scale, the prospects of complexity reduction, and of meeting externalist demands. Let me explain. First, to think of adult human beings as having a second nature is to think of their inner and outer make-up as the result of a determinate, stable and invariant structure albeit an acquired one. But as I observed in section 1 de88
terminacy, stability and invariance are temporal properties. If adults are to have a determinate, stable and invariant second nature this implies that changes in this structure do not appear over time during adulthood. This might be plausible - the idea being that the changes that occur in a human being’s internal structure during childhood are frequent whereas they are infrequent, or even absent, during adulthood. As a consequence we might think of a child as having a determinate, stable and invariant structure only if we restrict the child’s being an object of thought to its existence during one single day (or a week, perhaps). In contrast, however, the assumption allows us to think of an adult as having such a determinate, stable and invariant structure even if we extend the time-scale and think of the adult’s being an object of thought for its entire life. Stated as radically as this, the assumption seems rather controversial. Recent research on attitudinal and behavioural changes over the life-span, does indeed support the idea that changes in inner and outer make-up become much more infrequent during adulthood, but they are never absent. Obviously Adults have a history too. Complexity is the second issue in an attempt to think of variation in inner and outer make-up as determined by a being’s nature rather than its history. If it makes sense to think of childhood’s history as terminating in adulthood’s second nature, it should be possible to summarize the totality of history’s determining effects in the characteristics of the resulting second nature. The implications of this claim are as follows. On the one hand it should be assumed that during childhood the individual’s history, rather than its complex nature, provides epistemologically better (or perhaps merely psychologically more economical) resources for understanding, predicting and anticipating the individual’s behaviour. And on the other hand it should be assumed that during adulthood the individual’s second nature, rather than its history, provides these more favourable resources. This seems plausible only if an adult’s second nature is less complex than its history, that is, if this second nature can be reduced to a limited number of habits, virtues and/or principles of rationality. I doubt that this can be done, particularly in contrast to the related assumption that this cannot be done in the case of a child’s nature. The third issue is externalism, in both its diachronic and synchronic form. If the child’s trajectory through its environment over time plays a crucial role in the determination of the child’s behaviour and also in the formation of the adult’s second nature, this seems to provide a role to play for both synchronic and diachronic externalism with respect to the child’s behaviour but merely for diachronic externalism with respect to the adult’s second na89
ture. That is, the determinants of the child’s behaviour can be relational properties of the child, properties that depend for their individuation, and thereby for their determining efficacy, on features of the child’s present, past and future environment. Think of simple examples: the child’s attempt to grasp a certain object requires her to keep her fingers and thumb at a certain distance from one another and also requires her to use an appropriate amount of strength. Which distance and how much strength is determined by the actual object (an apple, say, or an egg, or a cuddly toy) she tries to grasp (synchronic externalism), or alternatively by past or anticipated future experiences of grasping the object (diachronic externalism). The adult’s assumed second nature is an acquired nature, the result of education. Its determinants will, therefore, be relational properties of a diachronic kind. That is, the adult’s second nature will have properties that depend for their individuation (and therefore for their determining efficacy with respect to the adult’s behaviour) on features of the child’s past developmental trajectory. These properties would not be properties of the adult’s second nature if the child s/he was before would not have gone through a certain trajectory. These properties would not even be the properties they are if the trajectory had not been the trajectory it is. It is unclear and controversial as to what this externalism exactly amounts to. Does it suffice for a specific property of an adult’s second nature (say its mode of attachment) that it bears a merely causal relation to the relevant features of the developmental trajectory that terminated in the adult being, say, avoidantly attached? Or is it required for the property to be the property it is (say a fear for spiders) that it bears some kind of intentional (interpretive or meaningful) relation to the relevant features of the developmental trajectory that terminated in the adult having arachnofobia? This is a hotly debated question in the philosophy of mind (Burge, 1986; Hurley, 1998; Wilson, 2004) that I shall have to by-pass here. For now it suffices to observe that the line of reasoning in which education is related to maturity as its completion, gives a very different role to play to diachronic as opposed to synchronic externalism. The observation is rather paradoxical. The idea seems to be that to understand an adult’s second nature (in terms of which we are supposed to understand the adult’s behaviour) we need not investigate synchronic relationships between the adult and its present environment, but merely diachronic relationships between the adult and its past environment. But if this is to be plausible, we should be able to explain why synchronic externalism is an issue in childhood (it has to be in order for a history to be build up), but would stop being an issue in adulthood. I don’t 90
see how this could be done. Why would external influences fade out in reaching maturity? Possible answers might refer to popular but also controversial ideas about critical developmental periods. Referring to such ideas, however, is in the present context seriously problematical, due to the necessity to accept, and to continue to accept, the relevance of diachronic externalism. The adult’s second nature, after all, is to be understood as the result of a specific past trajectory. That is, external influences are not simply absent during adulthood; they remain crucial in that some of the second nature’s properties depend for their individuation (and thus for their determining efficacy) on the past environment of the adult in question. Only the synchronic externalism fades out. I suspect that the very idea of a second nature as indeed a nature (a determinate stable and invariant structure) produced in history by education and completed in maturity is incoherent. I shall not elaborate on this here, but merely note that I think the idea is basically a blurred mixture of two lines of reasoning that might both be coherent. We have seen these lines earlier in this section. One line of reasoning is to side with a thoroughly naturalistic developmental psychology, according to which second nature does not differ intrinsically from human nature simpliciter, but is merely a specific instance of one of the many (but not infinite) variations nature allows human beings to flesh out. On this line of reasoning nothing much hinges on the use of the adjective ‘second’, nor is there a deep distinction between things with a history and things with a nature. Things with a history are on this account simply things with a nature, although this nature is more complex and allows for a more or less great array of variations due to developments over time in various environments. The other line of reasoning accepts a stronger notion of history as an anthropological constant by denying a view of education as a process that can be terminated in maturity. According to this line of reasoning the idea that education is an appropriate response to history is incomplete and needs to be supplemented by a view of education as itself a historical process, i.e. as itself producing history. In the next section I shall discuss this line of reasoning.
91
Education as producing history Two features of education as an anthropological constant should be highlighted to understand the claim that education produces history. These are the role of design, or, better, of engineering and the role of rationality. Let me say just a bit more about each in turn, revisiting themes from Frieda Heyting’s insightful paper on pedagogical intentions and pedagogical efficacy (Heyting, 1992). Engineering is to be associated with intentionality and functionality. If the older generation engages in downstream epistemic and moral engineering they take efforts to reach an aim, an aim that cannot be thought of as intelligibly related to these efforts unless there is a functional assumption, i.e. unless there is some idea of the educational setting as a system that can be described in terms of the causal roles for the whole of the possible (psychological and social) states of its parts. Heyting expresses this as follows: “...in pädagogischen Intentionen wird Wünschenswertes in bezug auf die miteinander verbundenen psychischen und sozialen Systeme vorausgesetzt und jeweils verknüpft met der Überzeugung, darauf ausgerichtetes Handeln sei sinvoll und möglich” (1992, p. 139). If education means for the older generation that they partake in downstream epistemic and moral engineering, this means that they should try to aid the young to acquire and develop relevant epistemic and moral capacities with more speed and success. Just as Heyting emphasizes, this requires not merely a functional setting and an intentional attitude on the part of the older generation. Education is not merely engineering, but it is a specific type of engineering, one that necessarily takes recourse to the normative structure of rational interaction. That is, education is normatively oriented. Education is directed at improving the quality of the younger generation’s attuning to their environment. In Heyting’s words, who takes into account that attuning between the young and their environment is not merely a matter of assimilation but should allow for empancipatory participation: I maintain that educational discourses fulfil a specific reflexive function in society, pertaining specifically to whichever kind of ‘better’ participation from new generations we strive for in the specific kind of ‘better’ future society that appears desirable to us. (Heyting, 2001, referring to Heyting, 1992) In my terminology this means that in education the older generation aims to support the younger generation’s efforts to respond appropriately to environmental requirements, and also to support the younger generation’s efforts to care for the appropriateness of their own responses. As I have argued 92
elsewhere (Bransen, 2006), a specific class of self-regarding reasons should be understood as belonging to the class of environmental requirements. This should suffice to take into account Heyting’s sympathetic and plausible observation that the young should not only assimilate to their environment as a given, but should be encouraged to participate in its determination (cf. Heyting, 1997). There is independent support for this in the very idea of the normative structure of rational interaction, support that is crucial to understanding how education produces history. Here is the idea. Rational interaction, i.e. the giving of and asking for reasons, presupposes at least two interlocutors, each of which should be considered by all of them as in principle co-authoritative about and co-responsible for the import of the rules that regulate their attuning to one another and their environment.3 In terms of an example, if I give my son a reason why he should obey me, and if I want him to obey me for this reason, I accept - precisely in virtue of my attempt to rationally interact with my son - that my son has a say in determining the import of this reason. Of course, his authoritativeness in this matter is not just a question of his whim, but is itself a consequence of his attempt, if there is such on his behalf, to rationally interact with me. Much follows from this, among others the idea that education produces history. Here is why. In education the older generation tries to improve the younger generation’s capacity to attune to their environment. This is the capacity to interact rationally, where this should be understood in a very broad sense that allows for instance for emotions, among many other things, to be reasons, and that accepts that there may be many and enduring impediments to the explication of the reasons we feel to have. So this is not a plea for an over-rationalistic conception of educational arrangements. The only crucial feature of the picture is that the attuning between human beings and their environment is a normative issue, and that human beings are therefore right in having an interest in getting the import of the relevant norms right. But getting the import of the relevant norms right is a historical process. It takes time, filled with interactions between particular human beings. Norms are not simply given features of a natural environment. They require justified expectations, and these require, for their existence to be possible at
3
I have said more about this on a number of occasions. See Bransen (2001, 2002, 2003). What I have said on this is deeply influenced by Pettit (1993), Pettit & Smith (1996), Brandom (2000).
93
all, social beings with intentional and authoritative states of mind.4 Norms are, therefore, a prime example of things with a history, things with properties that are determined by a specific trajectory through time, a trajectory that necessarily involves a process of interacting human beings that invest their expectations, their sensibilities, expertise, authority and power to determine the import of all the relevant norms that will regulate their attuning to one another and to their environment. Now we can see what it means that education produces history. Downstream epistemic and moral engineering takes place on billions of occasions among billions of people, and if getting the import of the relevant norms right constitutes a substantial portion of the interactions between the young and the old, on each occasion, it is completely inconceivable how this could fail to be an intrinsically historical process from which emerges a great variety of different norms. On a global scale downstream epistemic and moral engineering produces a multitude of norms, each determined by its history, rather than its nature, i.e. each determined by a particular series of interactions between a limited number of people from different generations. It is these norms that are historical, produced over time by education, and it is these norms that we need to understand human behaviour. These norms don’t make up an adult’s second nature. As years go by human beings may develop a more insightful, more reliable, and psychologically speaking also more rigid grasp of the import of the norms that regulate their attuning to their environment. But the resulting stability is not a matter of overcoming history, of reaching a state of maturity conceived of as the achievement of the end of education, a state of nature, second nature. There is no end to education, and thus, no end to history.
Education without maturity I should like to conclude by returning to the main question of this conference, and to use the results of the preceding discussion to explain why we could have an interest in answering this question but also to provide some support for my claim that the question is ill-conceived.
4
Cf. De Waal 1996 for a story about how this history might have taken place during the evolution of mankind.
94
I began this paper by providing an obviously correct and trivial answer to the central question of this conference: the new should conform to the old when they are right, and should oppose them when they are wrong. A striking feature of this answer is that it completely ignores the key terms of the question. Whether or not the young should conform to or oppose the old has, according to this answer, nothing whatsoever to do with the fact that the young are young and the old old. The correct and trivial answer emphasises that the discriminating factor between conformity and opposition is rightness, not generation. But if that’s the correct answer, why should we have posed the question as if it had something to do with generation? Why should we have expected there to be a general answer concerning the appropriate attitude of the younger to the older generation? Why should we have expected it to be the case that the asymmetrical distribution of power, authority, and expertise among the different generations should lead to an asymmetry in the direction of rational interaction? I’m inclined to think there is an underlying, implicit assumption at work here, not a pragmatic presupposition with a foundational function, but an illconceived assumption that power, authority and expertise on the one hand, or unbiased, original sensibility on the other hand, are somehow analytically related to epistemic and moral rationality. That is, if power, authority and expertise were analytically related to rationality, it would make sense to think that the older generation will as a matter of course be right about the rules that regulate human beings’ attuning to their environment. Their rightness will provide support for the claim that the young should, normally speaking, conform to the old. But if on the other hand unbiased, original sensibility were somehow analytically related to rationality, it would make sense to think that the younger generation will as a matter of course be right about the rules that regulate human being’s attuning to their environment. On this line of thought the young should, normally speaking, oppose the old. On the basis of the argument developed in the previous two sections, it might seem plausible to maintain that the underlying assumption discussed here is related to the idea of education as terminated in maturity. That is, the assumption might seem to be related to a very serious and substantial distinction within mankind between the generations. According to this distinction, children have a history rather than a nature, but adults have overcome history in virtue of their second nature. As a consequence, there should be a general answer to how these distinct kinds of human beings should attune themselves to one another. And then it depends on whether one thinks of history as primarily an emancipatory power of mankind, or rather as primar95
ily a source of unintelligibility, unpredictability and confusion, whether one is inclined to side with the young and preach opposition or side with the old and preach conformity. It will be rather obvious now that my argument in the previous section provides most support for a rejection of the underlying implicit assumption. There are no two kinds of human beings, historical ones that are young, and sensible, but unpredictable and confused as well, and post-historical ones that are mature, and experienced, and in possession of a second nature. There are just educatable persons. They differ of course hugely, depending on whether they are old and experienced or young and exploring. But they don’t differ in their historicity. They are all exemplars of homo sapiens, exemplars of human nature, best conceived of as running through history, in virtue of which each new exemplar displays an inner and outer make-up that is determined by the specific trajectory of this individual human being through its environment over time during its entire life-span. It is our concern for the import of history on our lives in conjuction with a mistaken conception of education as terminating in maturity, that explains our interest in the central question of this conference. Without this mistaken conception of education our concern for the import of history on our lives will not weaken. But we will be able to understand this concern better - as a concern for human life as an experiment we should like to endorse without being entitled to claim that we know how it will succeed.
References Brandom, R. (2000). Articulating reasons: An introduction to inferentialism. Cambridge MA: Harvard University Press. Bransen, J. (2001). On exploring normative constraints in new situations. Inquiry, 44(1), 43-62. Bransen, J. (2002). Normativity as the key to objectivity: An exploration of Robert Brandom's articulating reasons. Inquiry, 45(3),373-392. Bransen, J. (2003). De opvoedbare mens. In M. van Hees, E. de Jonge & L. Nauta (Eds.). Kernthema's van de filosofie (pp. 89-109). Amsterdam: Boom. Bransen, J. (2006 forthcoming). Selfless self-love. Ethical Theory and Moral Practice, 9. Burge, T. (1986). Individualism and psychology. Philosophical Review, 95, 3-45. Cassirer, E. (1923-29). Philosophie der symbolischen Formen. Darmstadt: Wissenschaftliche Buchgesellschaft.
96
Cassirer, E. (1944). Essay on man. New Haven: Yale University Press. Gopnik, A., Meltzoff, A. & Kuhl, P. (1999). The scientist in the crib: What early learning tells us about the mind. New York: Harper Collins. Heyting, G. F. (1992). Pädagogische Intention und pädagogische Effektivität. In N. Luhmann & K.E. Schorr (Eds.). Zwischen Absicht und Person; Fragen an die Pädagogik (pp. 125-154). Frankfurt/Main: Suhrkamp. Heyting, G. F. (1997). Het vanzelfsprekende en het discutabele : een schets van opvoedkundig grondslagenonderzoek. Inaugurele rede. Amsterdam: Universiteit van Amsterdam. Heyting, G. F. (2001). Antifoundationalist foundational research: analysing discourse on children’s rights to decide. In G. F. Heyting, D. Lenzen & J. White (Eds.) Methods in the philosophy of education (pp. 108-124). London: Routledge. Hurley, S. L. (1998). Vehicles, contents, conceptual structure, and externalism. Analysis, 58, 1-6. Kant, I. (1784). Beantwortung der Frage: Was ist Aufklärung? Berlinische Monatsschrift, Dezember-Heft, S, 481-494. Margolis, J. (1993). The flux of history and the flux of science. Berkeley: University of California Press. McDowell, J. (1994). Mind and world. Cambridge, MA: Harvard University Press. Pettit, P. (1993). The common mind: An essay on psychology, society, and politics. Oxford: Oxford University Press. Pettit, P. & Smith, M. (1996). Freedom in belief and desire. Journal of Philosophy, 93(9), 429-449. Sartre, J.-P. (1946). L'existentialisme est un humanisme. Paris: Editions Nagel. Sterelny, K. (2003). Thought in a hostile world: The evolution of human cognition. Oxford: Blackwell. Waal, F. de (1996). Good natured: The origins of right and wrong in humans and other animals. Cambridge, MA: Harvard University Press. Wilson, R. A. (2004). Boundaries of the mind: The individual in the fragile sciences. Cambridge: Cambridge University Press. Wittgenstein, L. (1953). Philosophical investigations (G.E.M. Anscombe, Trans.). G.E.M. Anscombe and R. Rhees (Eds.). Oxford: Blackwell.
97
8 Education from the inside and the outside. Response to Jan Bransen JUDITH SUISSA
I would like to start by agreeing with Bransen’s conclusion that the question “should the young obey the old?” is misconceived. Yet en route to this conclusion, Bransen highlights particular aspects of education and history as “modes of being human” or “anthropological constants”, and I want to raise some questions about this approach. Firstly, in talking of education in this anthropological sense, Bransen implies a spectator’s view of the educational process. Secondly, and connectedly, his talk of history focuses on a trajectory of accumulated knowledge and experience, rather than on notions of meaning. Bransen formulates his rejection of the question “should the young obey the old?” in terms of epistemic notions – arguing that the trivial answer that we should obey the old if they are right and disobey them if they are wrong is brought into question by his analysis. Yet I suggest that this perspective obscures a crucial aspect of what education means. In effect, I believe Bransen’s discussion confuses two senses of the word “education”. “Education” is sometimes used to refer to the process more accurately described as socialization, and it is this sense that actually informs most of Bransen’s discussion. Of course it is true to say that education as socialization involves a considerable degree of what Bransen refers to as “downstream epistemic engineering”, by which process the older generation assist the young in “acquiring and developing relevant epistemic and moral capacities”. This is the process that goes on, often unintentionally, and in every society we can imagine, alongside and often before any formal, intentional process of education or schooling. It is what goes on when, unthinkingly, I use a particular linguistic register, vocabulary and tone in speaking to my 2-year old child; it is what goes on when pre-school children observe the behaviour of adults at dinner-parties and in supermarkets; it is what goes on when we include our children in various festivals and rituals, and when 99
they begin to read books and watch films. This sense of education as socialization is, however, best grasped not through the contrast between education and history, but that of education and culture. Whether culture is taken to mean the specific culture of an ethnic or national group, a social class, or a universal human culture, the point is that the process of socialization is a basic requirement for children to be able to function in society. Someone in whom this process has failed could be described as uneducated, but it would be more accurate to describe him as not socialized. Thus the newborn child who, as Bransen says, seem to us “the most disturbing as well as the most vulnerable stranger” and whose behaviour is “unintelligible, both unpredictable and confused” is so, I think, not because she does not have, in Bransen’s words, “a history”, but because she does not have a culture – in the widest sense of human culture. This point is even more evident when one considers extreme cases such as Victor, the so-called “Wild boy of Aveyron.” In his work with Victor, Dr. Itard painstakingly tried to replicate, through systematic training, the unreflective socialization process usually performed by adults with newborn children. Victor did not lack a history: but the trajectory of his history had taken place in a non-human social group; he therefore appeared to lack a human culture. However there is another sense when we use the word “education” to mean something other than socialization, and it is this sense which I think Bransen obscures in his paper due to his focus on the anthropological – or spectator’s – perspective. In this second sense, when we talk of something as being “educational” we are referring to something more than a successful process of “epistemic engineering”. What is this something? We can all, surely, make sense of the idea that we may have gone through processes designed specifically, and in good faith, to “educate” us, but have not found these experiences “educational”. What defines something as educational in this sense? Philosophers have notoriously tried to answer this question with varied degrees of success, but I think there is some intuitive truth in the idea that to understand something as educational in this sense demands that we focus on questions of meaning rather than of rightness. Yet the question of whether something is meaningful or not and to what extent it is, cannot be answered from an external, anthropological point of view, but only from an inner one. Crucially, while the anthropological sense allows us to trace a linear course of a person’s educational development from the outside, as it were – and thus to determine to what degree she is or is not socialized – this second sense of education often 100
does not appear to be linear. Indeed, as Jane Roland Martin has argued, it often proceeds in unpredictable and transformative ways. “Everyone knows that even the very best education can be an unruly affair and is almost always an unpredictable one.” (Martin, 2002, p.9) Education in the socialization sense in which Bransen uses it is, in contrast, often a pretty predictable affair. We can fairly safely predict that a child brought up by middle-class, English-speaking parents in Britain will, without much conscious planning on their part, turn out to be a middle-class English-speaking Brit. But apart from these external labels, the success – or failure – of the socialization process tells us very little about who she is as a person. She has a history, in Bransen’s sense, but this is something that is only seen from the outside. What is left out is the inner world. The trajectory of her life may be to a large extent predictable, and thus understandable; knowing it will allow a stranger to interact with her fairly easy on a certain level. As I suggested earlier, I believe that this has far more to do with culture than with history, and find it puzzling that the word “culture” is entirely absent from Bransen’s discussion. Yet the crucial point here is that however much we know about this person’s history and trajectory of socialization, we can never predict or even know the meanings that she will come to make of this trajectory; of the events and incidents thrown at her. Although it may be true of education as socialization that it proceeds along a linear path – we cannot suddenly turn into educated members of society – it is often true of this second sense of education that it proceeds, as Martin suggests, by a series of transformations. Looking from the outside at the metamorphoses that shaped the lives of the people Martin discusses - people like Malcom X, for example – we could perhaps, with enough information, analyze each transformation in terms of an unbroken series of changes. But the crucial point is that “each metamorphosis is more than the sum of its parts”. I want to suggest that what accounts for this “something more” is the narrative that we, as individuals, construct for ourselves. As MacIntyre (2004) suggests: “I am the subject of a history that is my own and no one else’s, that has its own peculiar meaning” To bring out this contrast, let’s look at a famous quote that ostensibly demonstrates one of the possibilities suggested by the guiding question of this conference. Einstein once said: “If I have seen farther than others, it is because I was standing on the shoulders of giants.” Now we could read this statement from an external, anthropological position, implying that we can understand Einstein’s unique insight in to the working of the universe by 101
construing his education as a trajectory leading from Newton, through modern physicists and early educational influences. But while this may be a useful idea in the context of an essay on the philosophy and sociology of science, or possibly on the psychology of genius, it tells us little about Einstein as a person. Einstein’s contention that he stood on the shoulders of giants is not a story of causality, but a narrative of meaning and identity. I am not implying that Bransen ignores this sense of education. Indeed, in arguing, as he does towards the end of his paper, that “there is no end to education and thus, no end to history”, he is, I believe, using education in this second sense. But the dominance of the anthropological model obscures the essential sense in which education involves an inner, as well as an outer perspective. Likewise, education as the construction of meaningful narratives involves a far more dialogical relationship with those doing the educating than the “downstream epistemic engineering” which Bransen refers to. In emphasizing the way “the older generation tries to improve the younger generation’s capacity attune to their environment” (p. 93) – Bransen is focusing on the perspective of the educator or the neutral observer, and of course this too could be construed as part of a narrative – comprising the educator’s intentions, desires and aims. Yet whether or not a particular educational encounter counts as truly educational depends at least in part on the pupil’s own narrative. For education in this sense, although concerned with “the inner”, is not purely subjective, but it is of course inter-personal. In keeping with Bransen’s conclusion that there is no end to education, we are continually reweaving our narratives, and this is partly a matter of the particular ways in which we can make use and sense of other narratives within our own. Bransen talks of “the young” and “the old” as anthropological categories. But in any educational encounter, one is never dealing with “the young” and “the old”, but with individuals engaged in constructing their own narratives as well as intersecting and culturally shared narratives. The question of whether the young should learn from the old, then, only makes sense if we ask who these people are, and what kind of an educational relationship they find themselves in. If I encounter teachers – perhaps older than me, or perhaps, as in the case of any philosophers, dead – who can help me to construct a meaningful narrative, this is truly educational. Perhaps the only duty of the young and the old, then, is to show a mutual willingness to engage in such encounters. Bransen, in mentioning the question as to whether a property of what he refers to as an adult’s second nature should be understood as bearing a 102
merely causal relation to the relevant features in the developmental trajectory, or whether it should be understood as bearing “some kind of intentional (interpretive or meaningful) relation” to them, decides to “bypass this hotly debated question”. But this, surely, is the question in education; if education is to mean more than socialization, or the accumulation of knowledge and skills, it must, surely, be about meaning? Talk of influences and causality does not get close to capturing this sense. “There are no two kinds of human being”, Bransen concludes: “just educatable persons”[..], exemplars of human nature, but best conceived of as running through history, in virtue of which each new exemplar displays an inner and outer make-up that is determined by the specific trajectory of this individual human being through its environment over time…” Yet it is exactly this conflation between the inner and the outer make-up that needs to be pulled apart, I suggest, if we are to develop a fuller account of what education means. And although the anthropological account goes some way to answering the question of this conference, we need the other part as well. For as Mary Midgley (2005, p. 344) reminds us: “These two aspects of life are not two kinds of stuff or force. They are two points of view – inside and outside, subjective and objective, the patient’s point of view on his toothache and that of the dentist who studies it. The two angles often need to be distinguished for thought. But both of them are essential and inseparable aspects of our normal experience…”
References MacIntyre, A. (2004). After Virtue. London: Duckworth. Midgley, M. 2005. A Plague on Both Their Houses. In D. Midgley (Ed.). The Essential Mary Midgley. London: Routledge. Roland Martin, J. (2002). Educational Metamorphoses. The George Kneller Lecture. American Educational Studies Association.
103
9 The student Klaus Mann. Betwixt and between individualism and “generation-unit” BRITA RANG
I A reawakening of interest in the ‘concept of generation’ has taken place within the social sciences (see f.i. Edmunds & Turner, 2002; Elder 1999; Kohli, 2003; Kohli & Szydlik, 2000; Turmel, 2004; Vincent, 2005). The essay by Karl Mannheim – published in German in 1928, appearing as The Problem of Generations in 1936 – is still one of the fundamental works, in the discussion of the potential of this concept. But critics (e.g. Bartels, 2002, p. 1) suspect that the term generation counters the individualisation of biographies with the topos of “Gemeinschaft” (Bartels, 2002, p.14). Sociologist Kohli points out that ‘generation’ was already used in the past as a term promising increased insight, which did not however hold true. According to him this had to do with problems inherent in the concept. One of these problems is the move from the continual succession of generations within primary groups (generation as a set of familial relations) to social generations (generation as a cohort with community of experience) (Kohli, 2003, p. 10). Others (e.g. Bude, 2000) state that current interest in the concept of ‘generation’, is bound to the analytical weaknesses of several other sociological terms, such as social class. To the contrary, the term generation offers the possibility to look at both, individual and collective processes, in a specific contingent socio-cultural situation (cf. Kohli, 2003, pp. 10). Mannheim expressed similar thoughts when he proposed that by using the concept of generation not only individual but also shared group interests could be articulated and actively represented. For this, however, generation as a ‘location’ needs to be distinguished from generation as an ‘actuality’: “Whereas mere common location in a generation is of only potential significance, a generation as an actuality is constituted when similarly ‘located’ contemporaries participate in a common destiny and in the ideas and concepts which are in 105
some way bound up with its unfolding. Within this community of people […] there can then arise particular generation-units. These are characterized by the fact that they do not merely involve a loose participation by a number of individuals in a pattern of events shared by all alike though interpreted by the different individuals differently, but an identity of responses, a certain affinity in the way in which all move with and are formed by their common experiences” (Mannheim, 1952 (1936), p. 306). Thus within any generation a number of differentiated, antagonistic generation-units can exist. Within the current debate, and also in Mannheim’s work, the concept of generation appears to be the result of primarily theoretical efforts, either within the scope of macro- or micro-level analysis. Here the concept seems to occupy an analytical position, which other terms could not fill in a substantiated way and which opens up new descriptive options. I want to emphasize however that this term was already an established descriptive category for everyday intellectual life in the 1920s, even before sociologists like Mannheim took it up.1 In the 1920s debate it became a category that could describe certain shared historical experiences especially of a youth that witnessed World War I. Within this concept of youth, a generation-unit was articulated that was closely related to developments of the youth movement and the pedagogical reform movement. When World War I provides the shared political-historical background, pedagogical images and arguments provide the crucial argumentative framework for this generation-unit. I want to elucidate this thesis with regard to the biography and generational experiences of Klaus Mann. Born in Munich in 1906 the second child of Thomas and Katja Mann, Klaus Mann articulated this image of a generation from a very early age and announced himself the spokesman for it starting in 1924. According to the editors of his early work, Klaus Mann wanted to “find a language for his generation, a ‘melody’, a meaning of life” (Naumann & Töteberg, 1992 , p. 13).2 Mann claimed to be speaking for a generation-unit, “because no generation had more right to desperately being taken seriously than ours” (Mann, 1992, p. 51).3 This is the very young Klaus Mann speaking. Later in an essay appearing in 1938 (Die Kriegs- und 1
Mannheim himself pointed to two analytical traditions of the topic ‘generation’: 1. the positivist formulation of the problem (Auguste Comte), aiming at a general objective law of social change; 2. the romantic-historical formulation of the problem (Wilhelm Dilthey), searching for the inner totality of an historical epoch. (Mannheim 1936/1952, 276 et seq., 280 et seq.). 2 “für seine Generation eine Sprache, eine ‘Melodie’, einen Lebenssinn zu finden”. 3 “keiner Generation (kam) dieses verzweifelte sich Ernst-Nehmen unbedingter, ja notwendiger (...) zu als der unsrigen”.
106
Nachkriegsgeneration), when he was already thinking very differently about the shared interests of (t)his generation, he was still describing the reality of such a unit: “that the people of one age-group are connected by certain experiences, certain vital suppositions and complex feelings, which are necessarily foreign in this form to younger and older people or may even be completely incomprehensible” (cf. Mann, 1969 (1938)).4 Below I will attempt to show in which sense this image of a generation was formed but also problematised by school experiences and developments of the youth movement.
II In his two autobiographies as well as in letters and several essays Klaus Mann has emphasised the great importance of a certain short period during his schooldays for his spiritual development. In German literature it is rare to find such a favourable appraisal of school. We may search in vain for such a positive evaluation in the work of Thomas and Heinrich Mann and in that of other authors belonging to Klaus Mann’s generation. It is not difficult to come up with an explanation for this apparent exception. Both schools Klaus Mann speaks so highly of were not traditional gymnasia (grammar schools). Klaus Mann found the requirements of the gymnasium in Munich uninteresting and its teachers bored him. The schools he attended were two ‘new education-type’ boarding schools. ‘Landerziehungsheime’ as they were known in Germany at that time. One was the ‘Bergschule’ in Hochwaldhausen near Fulda, the other the famous ‘Odenwaldschule’ in Oberhambach near Heppenheim. He went to the Bergschule, together with his sister Erika, for just a few months (spring - summer 1922). He attended the Odenwaldschule for a year (1922-1923).5 Nevertheless, he perceived this one year to have played 4
“...dass die Menschen eines Alters durch gewisse Erlebnisse, gewisse vitale Voraussetzungen und Gefühlscomplexe miteinander verbunden, die den älteren oder jüngeren in dieser Form notwendiger Weise fremd oder gar unverständlich sein müssen”. 5 Under the guiding hand of its founder and principal, Paul Geheeb, the Odenwaldschule (which first opened its doors in 1920) was one of the few ‘Landerziehungsheime’ to keep a fair distance from militaristic and nationalistic tendencies (cf. Shirley, 1992). This attitude was already apparent during the first world war, and was inspired by ethical motives, influenced by oriental philosophy. Geheeb attempted to create a ‘humane school’. His pupils came from a multitude of European and non-European countries. They belonged to different religions and cultures and were usually the children of very well-to-do families. Paul Geheeb tried, primarily indirectly, to instil his pupils with mutual tolerance, understanding of and openness
107
such a crucial part in his intellectual and social development, that in 1930 he wrote to Paul Geheeb, the founder and principal of the Odenwaldschule: “Dear Paulus, I can’t imagine my life without the year at the ‘Odenwaldschule’, although it wouldn't have had this effect on me had you not been there” (T. Mann, 1961, p. 69). In an essay written in 1930 he states: “one should re-experience the intellectual sensory indulgences of adolescence, and also the milieu in which these impressions occurred, in my case the atmosphere of the German youth movement, the free school community, into which I was introduced in my fifteenth year and which influenced my awakening consciousness in a decisive manner” (Mann, 1992, p. 334). With regard to those experiences he saw himself as defined by the similarity of the ideas, sentiments, and potentially actions of ‘his generation’. But just what fascinated Klaus Mann so much about these ‘Landerziehungsheime’ (boarding school) and above all about the Odenwaldschule in particular and to what extent did he experience there - what Karl Mannheim called – “a youth generation as actuality” (Mannheim, 1952, pp. 302)? To learn more about his life is not the only reason for asking this question. The person and his work concern me, of course, but besides this I am interested in the specific developmental processes which made Klaus Mann a typical example of a bourgeois intellectual within a generational unit “which work(ed) up the material of their common experience in different specific ways” (Mannheim, 1952, p. 304): what influenced his development, as he himself sees it, from ‘irrationalism’ to rationality, from a politically disinterested and individualistic artist into an anti-fascist and democratic socialist. In this context I am not primarily interested in the specific, concrete characteristics of the two ‘Landerziehungsheime’ where Klaus Mann stayed; I would also like to focus my attention on the changeable phenomenon of the youth movement as a whole (Klaus Mann too, experienced the youth movement as a typically German phenomenon) and the education reform movement, the new education (German: Reformpädagogik). In Germany these two ‘movements’ were closely interrelated (cf. Nohl, 1933; Scheibe, 1969, Oelkers, 1996). Important elements of both movements played a central role in the ‘Landerziehungsheime’ Klaus Mann attended. However, it was by no means towards others. And so in the twenties Paul Geheeb withdrew from his ‘völkisch’ colleagues from most of the other ‘Landerziehungsheime’. While they hailed the national socialist revolution in 1933 (see for example Andreesen, 1934), Paul Geheeb referred to the nationalsocialists as “a band of criminals at the top of the state” (cited by: Feidel-Mertz 1983, p. 29). In 1934 he emigrated to Switzerland with a few of his pupils, where he, with great difficulty, started his ‘École de l’humanité'.
108
always the case, or a matter of course, that the influence of the youth movement and the ‘Reformpädagogik’ had an immunizing effect against emerging National Socialism. It is, moreover, one of the historical ambivalences of the German youth movement and of a substantial part of the German 'Reformpädagogik', that here (!) too a partly direct, partly indirect affinity existed with National Socialism (cf. Rang & Rang, 1984; Oelkers, 1996). When Klaus Mann – as a writer – interpreted his impressions of the German ‘Landerziehungsheime’ in retrospect as a writer, he did it in the context of political developments and changes within society, thereby putting his impressions into perspective. Nevertheless, he remained faithful to what he considered to be constitutive for his formative process. So what was it that he found so impressive, and what troubled him so much about his education? The practical work in the kitchen (peeling potatoes), in the workshops and the garden did not appeal to him. He complained to his father about this damned ‘practical work’ (Mann, 1987, p. 2). He didn’t enjoy sports much either, although luckily for him sport didn’t play such a prominent role at the Odenwaldschule as it did at the ‘Landerziehungsheim Schloss Salem’, which Golo Mann was later to attend. Klaus Mann was more fortunate than his brother in this respect. In all probability the ... ‘Luftbad und Nacktkultur’ (open-air bathing and culture of nudity) came under the same heading. This was part of the morning ritual at the Odenwaldschule, as its teachers were convinced this would help to: “get used to nakedness ... one of the best antidotes for unhealthy curiosity and excitement” (Huguenin, 1926, p. 15). What fascinated the fifteen- and sixteen-year-old Klaus Mann was in fact something the educators had not intended, namely the effects of the “cohabitation of young people in perfect freedom” which was experienced as laden with erotic tension and as extremely inspiring (cf. Mann, 1976 (1942)). The excitement fuelled by this atmosphere inspired his first published literary experiments: Die Jungen (1925b (1923)), Der Alte (1992 (1925)), Anja und Esther (1925a). In Kind dieser Zeit (1932) and in Der Wendepunkt (see 1976 (1942)) he has a lot to say about this aspect of his experiences: “It seems to me now as if I lived in a constant state of utmost religious ecstacy” (Mann, 1965 (1932)). And: “we were all tremendously intense [...] everything stimulated us to hold excited conversations, discussions of a passionate exploring, groping, all-inclusive, bombastic, confused nature [...] Everything was geared towards proving one’s own geniality to oneself and one’s discussion partner” (Mann, 1976 (1942)). In Kind dieser Zeit (1932) Klaus Mann pays attention to the diverse forces which accompany this intensive process of growing up. At twenty-six he remembers his boarding school period as a 109
“mixture of spiritual ecstasy and gluttony”, of mysticism and of “visions of vice” (Mann, 1965 (1932)). His parents in Munich received frequent requests for tuck-boxes and especially for sweets. The scanty meals served at both boarding schools were obviously insufficient to satisfy this physiological hunger. The hunger for intensive literary, spiritual and philosophical discussions laden with erotic tension, on the other hand, constantly found food and satisfaction, although (or perhaps because) these usually took place outside the regular school activities. The philosophy of the new education granted pupils a certain amount of ‘free space’, and together with the pedagogical atmosphere of the remote boarding schools, isolated as they were from external influences, a fertile medium was indeed created for the satisfaction of this hunger. In this peculiar state of self-reliance and agreeable isolation – as if on a magic mountain – Klaus Mann develops his characteristic fusion between “the disappearance of different ages”, “strong vitality”, “destructive and negativistic pathos”, and “restless and tormented lust for sensation” (Mann, 1932, p. 128-129). The almost endless discussions on mystics, but also on Novalis, Nietzsche and Wedekind, send him into an “ecstatic state of self-consciousness” (ibid., p. 140). In a school essay he envisages himself and his peers as heirs to and successors of Nietzsche. In these memoirs the school is conveyed as a place in which “extreme unlimited individualism” (almost drove) the participants in the discussion “to selfidolisation” (ibid., p. 144). Besides such feelings of omnipotence the sixteen and seventeen year old also notes with remorseful self-pity: “Alas, what am I? My ecstasy is just inferior puberal excitement?” (ibid., p. 144). In comparison to these memories of the atmosphere generated by stimulating conversations, the lessons themselves seem rather dull. In a letter written from the Bergschule to his father on the 17th of June 1922 he writes: “We learn much too little here ... Our stay here is not only saddening, but also completely useless to us. I’m not saying the value of the school is negative, it’s just null” (T. Mann, 1961, p. 10). Klaus Mann wasn’t as critical of the education at the Odenwaldschule, which incidentally didn’t take place in classes of children of the same age group doing the same work, but was organised in the form of courses which the children followed depending on their prior knowledge. But even here the main source of influence and inspiration was not the course system itself, but rather experiences and conversations with other pupils of approximately the same age, as well as his own literary experiences, and reflections. At first Klaus Mann participates in two courses (philosophy of religion and theory of harmony). During the last few months of his stay at the Odenwaldschule he was exempt from everything of 110
a more or less compulsory nature. He can do as he pleases. “I didn’t have to do any of the courses or the practical work; I was allowed to amble, write poetry and meditate all day long” (Mann, 1965 (1932)). It would be incorrect to suppose that Klaus Mann received preferential treatment at the Odenwaldschule, either because he was Thomas Mann’s son, or because he was perhaps a ‘difficult’ pupil. It was one of the principles of this Landerziehungsheim to take into account the individuality of children and adolescents and to respect and meet the needs of developing individuals. In Paul Geheeb’s opinion the educator should take into consideration the interests the pupils had already developed prior to their entrance to the school and those developed during their stay. At the Odenwaldschule these were considered “profound and irrepressible instincts” (Huguenin, 1926, p. 20); Geheeb probably saw them as entelechies. He worshiped Goethe and made Pindarus’s principal ‘become who you are’ his educational creed. Where young people develop intense interests for certain things, problems or activities unbridled forces are at work. The educator’s task should not be to curb or correct, but to encourage and stimulate (ibid., p. 20). This wasn’t too difficult in the case of Klaus Mann, as he obviously showed interests the educators at the Oldenwaldschule considered valuable. Klaus Mann was refused a place at the previously mentioned élitist Landerziehungsheim Schloß Salem run by Kurt Hahn. This school was oriented towards the school community and achievement and although Klaus Mann showed ‘very profound intellectual interests’ he also ‘cultivates his affected awkwardness in practical matters with vanity’. But the educators who judged him so harshly at Schloß Salem were right in recommending the Odenwaldschule as a more suitable boarding school. Klaus Mann’s affected behaviour and one-sidedness were not important to Paul Geheeb; what counted were the boy’s ‘very profound intellectual interests’. So this is why Klaus Mann was freed from everything he would otherwise have more or less been forced to do. Education became self-education in an inspiring environment, together with like-minded young people. The educators gave Klaus Mann the rights of an adult, the right to pursue and explore his own interests and to develop his own life-style and working methods. Geheeb, tolerant as he was, trusted the instincts of his pupil. His tolerance and faith were so great that he was even prepared to wait patiently and intently. Still under Geheeb’s spell Klaus Mann lets the ‘Alte’ in Anja und Esther say: “Here the confusion of young people is given ample scope – the tormented and restless children of the Lord can do as they please, undisturbed” (Mann, 1925a). 111
This is by no means a realistic characterisation of Paul Geheeb’s pedagogical style, although it is at least an associative memory in which an echo of Geheeb’s person is audible. Geheeb was extremely hurt by the first version of Der Alte (cf. Mann, 1992 (1925), pp. 97-99). Supported by his father, Klaus Mann made an appeal to his ‘artistic freedom’, even so some elements of what the young Klaus Mann thought he could see in Geheeb were included in these ‘poetical’ transformations. This didn’t destroy the feelings of sympathy Klaus Mann felt towards the ‘real’ Geheeb. These endured, especially in the form of gratitude towards the educator who granted him almost perfect freedom during an important period of his development. By doing this he was to create a constellation, in such an unsuspected place as a school, which reminded him of his parental home in Munich: “In the most beautiful and cleverest way they let us be” (Mann, 1932). “They let us run wild a bit” (ibid., p. 55). What he remarked about his parents, Klaus Mann could have also said about the educators at the Odenwaldschule. And precisely this attitude was an important formative factor not only for the development of a self regulated individualism but also – if I understand Mannheim correctly – for the development of the generation-unit.
III Another aspect of these memories of the Odenwaldschule is somewhat more elusive. I am no longer speaking of gratitude, but of irritation and, at least indirectly, criticism. I don’t mean Klaus Mann’s criticism of the ‘gymnasium’ in Munich. I am referring to his irritation and detachment towards educational theory in general, and the educators at both boarding schools in particular. He almost always takes such an attitude. What was it then which estranged him so much as a pupil, and later as an adult? Firstly the ‘Gemeinschaftskult’ (community cult) that was typical of the youth movement and of large sections of German progressive education. It is true that what was known as ‘Gemeinschaft’ at that time varied in emphasis according to the different directions within the youth movement and ‘Reformpädagogik’. For some the idea was linked to national or even nationalistic visions, for others it was closely connected to notions of democracy. Under this heading a broad spectrum of anti-societal, right-wing radical, neo-romantic, and sometimes socialist motives were brought forward. What all these widely divergent views of ‘Gemeinschaft’ have in common is the image of an idealised 112
community in which an all-embracing future ‘form of life’ was anticipated; this could be composed of the ‘Volksgemeinschaft’ or the whole of mankind. But almost everywhere where the ‘Bünde’ (alliances) within the youth movement or the individual ‘new’ schools propagated and brought the idea of ‘Gemeinschaft’ into pratice, feelings and emotional experiences were emphasised: “The strongest, the supporting experience of the youth movement was the experience of ‘Gemeinschaft’” (Scheibe, 1969, p. 43). Did Klaus Mann share this ‘supporting’ experience, which is in some way the keystone of his generation-unit? Yes and no. In retrospect he referred to the democratic ‘Schulgemeinde’ (school council), that is to say the meetings for children and youth which took place at regular intervals at the Odenwaldschule; he also speaks of a committee comprised of pupils, a forum for discussion and decision-making in communal matters. It is, however, as a politically conscious adult that Klaus Mann rakes up this aspect of his experiences at school. As a sixteen-year-old, he didn’t experience the ‘Gemeinschaft’ embracing all people as decisive. For him small groups were more important. In his puberal egocentricism he sought inspiration, understanding and the confirmation of his own person. At the Bergschule and the Odenwaldschule he didn’t find this by means in the abstract community of all school members, but he was able to find it in the concrete, intimate friendship with chosen, seemingly congenial fellow-pupils. Such boarding schools were the first to facilitate the formation of this kind of friendship. It was not only tolerated, but was also (partly) intended by the educators. Even so the individualistic exclusivity that Klaus Mann associated with it was also at odds with the community ideology these teachers simultaneously advocated. Klaus Mann tended to reject this ideology, rather than submit to it. The distinct nature of his individualism became clear to him very early on. In Die Jungen (1925b (1923)) which was to some degree inspired by his experiences at the Bergschule, he lets the ‘professor’ severely criticise the older pupils: “Especially the older ones, he said as he looked ahead, will not abide by our idea of ‘Gemeinschaft’, they choose to oppose it in an objectionable sterile way” (Mann, 1925b (1923)). In 1932 he describes his situation at the Odenwaldschule with greater detachment; he speaks of a “a kind of narcissism which was at the same time conceited and fiercely remorseful” the likes of which he was never to encounter again. (Mann, (1965) 1932). With realistic detachment he attributes this exalted “not very sympathetic spiritual condition” (ibid.) firstly to puberty and adolescence, then to the specific environment of the Odenwaldschule. “It [this spiritual condition, B. R.] strikes me on the one hand as something typically youthful, on the other 113
hand I think it was characteristic for the time and place which favoured it” (ibid.). And so a clear link is established which the child-centred educationists tended to overlook or suppress. They wished to promote self-responsibility and unselfishness, and not the “confused triumph of the ego” (ibid.). When we take into consideration that ‘extreme excessive individualism’ – at least in the case of Klaus Mann – was one of the outcomes of the ‘Gemeinschaft’oriented pedagogy of the ‘Landerziehungsheime’, we may inquire in retrospect whether these were just the problems of a spoiled, bourgeois, author’s son. Perhaps we are also talking about the pedagogical and historically illusionary ideal of ‘Gemeinschaft’ (community), of which the ambiguity, no the falsity, became apparent in 1933 at the latest. Klaus Mann, to say the least, hinted at this critical question. It was definitely not his style to posit a rough link between the new educational and national socialist cults of ‘Gemeinschaft’. What he concluded, with hindsight, was more like the existence of a curious affinity between the excessive cult of ‘Gemeinschaft’ and exaggerated individualism – les extrêmes se touchent –: “Because the Free school-‘Gemeinschaft’ unintentionally breeds the most extreme individualism as it were, in spite of its cult of ‘Gemeinschaft’” (ibid.). The ‘official’ handling of sexuality by the educators was also experienced by Klaus Mann as a contradiction. The ‘unofficial’ erotic pubertal impulses and experiences obviously differed from the expectations and wishes of the educators. Geheeb adhered to the dominant viewpoint within the youth movement by advocating a ‘pure’ and ‘true’ relationship between the sexes. He could identify with the creed of the ‘Wandervögel’: ‘Stay pure and ripen’; at the same time he supported and practised the principle of coeducation, which was extremely controversial at the time (at most of the ‘Landerziehungsheime’ girls and boys were not educated together). Coeducation was thought to be a good preparation and practice for future relationships between the sexes characterised by mutual respect and a ‘healthy’, ‘comradely’ association. In his opinion the school could promote such a relationship. So Geheeb rejected the “unnatural distance from the opposite sex” (Geheeb, cited by Kurzweil, 1973, p. 39) as was the practice at the more numerous boy’s and girl’s schools. According to him the atmosphere at this kind of school resembled that of a hothouse, due to the very absence of members of the opposite sex (ibid.). He wished to prevent such an atmosphere by introducing coeducation. Moreover, for him it wasn’t a case of erotic-sexual tolerance, but of “purity” and sublimation. He wanted the relationship between the sexes to be purely “platonic” (ibid.). Geheeb hoped to 114
encourage enthusiasm in his pupils for “strict and courageous self-discipline, for a loving relationship with one’s spouse and for an ideal form of monogamy” (ibid.). Klaus Mann’s memories of the Odenwaldschule remind us of the hothouse atmosphere Geheeb ardently hoped to prevent. In Kind dieser Zeit he speaks of a “peculiar erotic cynicism, a glaring indecency” (Mann, 1965 (1932)). For Klaus Mann’s circle of friends this complemented the “religious ecstasy” mentioned earlier, and as such it played an important role in the pupils’ development. He discerns a similar link between asceticism and dissipation, between religion and sexuality, in his own literary work: “Besides heartfelt religious lyrical poetry my exercise-books from that period contain some ‘visions of debauchery’, which have grown since” (ibid.). These visions don’t interest me (furthermore they have never been published). What I do find intriguing is Klaus Mann’s precocious, tendentious, critical sensibility. Because of this he was able to point out irregularities and defects which most of the educationists attempted to cover up or to gloss over. Youths like Klaus Mann and his friends, who were intensely occupied by their own erotic and sexual desires, were confronted with a not in the least bit erotic ‘pedagogic eros’. On these grounds they made a plea for ‘purity’, in which there was no room for their own sexual and erotic wishes and experiences. The young Klaus Mann is able to see the distortion, and the exaggeration characterising the ‘purity-postulate’ much more clearly than the adults, who were themselves the helpless victims of their own repressed defensive needs. He discerns its contradictory and even dishonest nature. He understands or suspects that the moralising platonism conceals denied sexual wishes and longings, especially those of his educators. On the basis of this suspicion, in connection with projections of his own wishes and longings, he carries out a telling transformation in the sketch Der Alte, first published in 1925 (cf. Mann, 1992, pp. 97-99). Not Paul Geheeb, but a caricature of him is portrayed. Not the champion of ‘pure’ human relationship, but a voluptuous, almost licentious ‘Alte’. The girls at the boarding school are not close to his heart in a pedagogical-metaphorical sense, but literally, as objects of desire. Geheeb was hurt, and he had reason to be. Although Klaus Mann defended himself by appealing to his artistic imagination (cf. the letter of 16-51925 in T. Mann, 1961, p. 19 ff.), this short story is surely also an expression of his critical feelings towards an idealistic, pedagogical moralism, which in spite of its contacts with the young remained very remote. The same theme: the discrepancy, as experienced by Klaus Mann, between the erotic experiences of youths and the official pedagogical rhetoric, 115
recurs in the narrative Die Jungen, which was also published in 1925 but which dates back to his ‘Landerziehungsheim’ period. This time the head of the school, ‘the professor’, lives in fear, and suspects that his pupils are Janus-faced, that they are not only ‘pure’ but also ‘impure’. Reality is therefore more complicated than the pedagogical ideal would suggest: “And for the second time this evening the professor takes a fright. He is fear-stricken and his breathing is laboured. What face is this, he thought, out of breath [...] What sort of pupils do I have, in God’s name? [...] He gets hot and giddy. What have I brought about, he pondered, what has become of the school I dreamt of? A strange and dangerous mixture of a monastery and a brothel” (Die Jungen, 1925b). In Anja und Esther this “mixture”, which the adults seldom dared look straight in the face, and even then only with ‘fear’ and aversion, appears once more. The “sanatorium for fallen children” (Mann, 1925a) is a place where “pious singing and dancing” form the setting for extremely “tragically intricate” erotic sexual relationships (Mann, 1976 (1942)). For the children and adolescents staying here the weird contradictions and ambiguities seem familiar and self-evident. Only Erik, an intruder from a faraway and totally alien world “out there” (Mann, 1925a) is surprised, confused and fails to understand: “Everything, the institute itself, is strange too. I’m at my wit’s ends. Whether it is a reform school or a cabaret, a kind of monastery - or the opposite - it all leaves me in an utterly confused state” (ibid.). What is the quintessence of the educator-figure and the corresponding self-image of the youths Klaus Mann speaks of? The ‘Alte’ in Anja und Esther lives and works in the middle of the oppressive ‘hothouseatmosphere’ (Geheeb) of his ‘sanatorium for fallen children’; at the same time it is as if he is miles away, hovering high above it, so to speak. Understanding and forgiving, and looking on from a distance, he lets the seemingly inevitable happen. He knows ‘everything’ (Mann, 1925a), but does not intervene. His highest maxim is the principle of ‘Wachsenlassen’ (letting grow), that of ‘laissez faire, laissez aller’. Klaus Mann was more than satisfied with the effects of this pedagogical stance in its intact form, namely that which predominated at the Odenwaldschule, and which the principal, Paul Geheeb, tried to pass on to his pupils. We already know that this school and its educators granted Klaus Mann ample licence to manage his own life, they ‘let him grow’. The situation described in Anja und Esther differs, however, from his own. The ‘Alte’ is a somewhat comical figure. Furthermore Klaus Mann keeps his distance by choosing the perspective of four young people: Anja, Esther, Kaspar and Ja116
kob. It is through them we are able to discover a motive of the utmost importance within the German youth movement, which is at the same time an extremely problematic one. We are referring to the characteristic craving for ‘leaders’. This longing was indeed consistent with a critical ‘turning away’ from the adult world. The growing independence of ‘youth culture’ is denoted by the catchphrase ‘Youth must be led by youth’. In at least one passage in Anja und Esther Klaus Mann expresses this desire for leadership. First he lets Esther articulate how the ‘youths’ keep the grown-ups at a distance: “In their unequalled scrupulousness they bring us into the world. They, indeed, trod on firm ground, which we had to lose. They had their little grievances, in which to nestle; their little weaknesses to take delight in” (Mann, 1925a). In short: the adult generation is both established and selfsatisfied: “But we have to feel threatened for hours upon end, such as no generation before us’. We stand there helpless between extremes and no one is our leader” (ibid., italics B. R.).
IV In 1925 and previously, Klaus Mann was part of a youth mentality characterised by disorientation (“standing helplessly between extremes”) instability (“We go on, without knowing where, uprooted and with all options open” (Mann, 1992 (1926), p.60), and the longing for a ‘Führer’. Within a year, however, Klaus Mann had developed a political conviction which was democratic, humanist and unmistakably and consistently anti-fascist. Large proportions of the youth movement and the German ‘Reformpädagogik’, not only forcedly but also of their own free will, gave their approval to national socialist developments. In conclusion I would therefore like to ask just what it was that enabled Klaus Mann to develop in this way? Why didn’t he become a hanger-on, following the path to the ‘Third Reich’ as most of ‘his’ own generation-unit? His original position would seem to offer favourable conditions for participation without resistance, being unpolitical, and one of aestheticism, mysticism and scepticism towards the claims of reason, spirit (German: Geist) and the intellect. However, towards the end of the twenties, at the latest, Klaus Mann began to perceive this choice more and more lucidly as a dangerous one. This marks the beginning of his own transformation from playful unpolitical aesthete, who - together with his friends adopted “eroticism and religiousness” as his “basic experience” (Mann, 1992 117
(1927), p. 131) into a politically thinking and acting intellectual, “who knows the meaning of concepts such as freedom, progress and democracy” (Mann, 1968 (1938)). One of his most differentiated essays written during this period of transformation (Mann, 1992 (1927), pp. 131-152) may be read as an early document of this transformation. The obsessions Klaus Mann demonstrated during his stay at the Odenwaldschule and which persisted up until a few years afterwards, are still manifest. He speaks of his “distrust of ‘Geist’, which had gradually become very strong”, in the present tense. Whether it found expression in the form of a malignant aversion towards ‘Geist’ or a saturation with ‘Geist’, this distrust clearly and unmistakably characterised the whole generation. It is “deep-rooted” (Mann, 1992 (1927), p. 25). Fuelled by an interest in the history of ideas, not by an interest in social history, he looks back upon one of the philosophical sources of this widespread aversion: “The only one thing we hated was rationalism. We had attempted to love all life has to offer, like the way Bergson’s philosophy recommends. This meant ‘surrendering to the secret, surrendering to stream and movement, and no longer to criticism. Because according to Bergson ‘the intellect is characterised by a natural lack of understanding of life’” (ibid., p. 35). Klaus Mann was no longer prepared to surrender himself totally to this irrational vitalism. At twenty-one he had reached, but not yet passed, his first ‘turning point’. He seeks and finds new references and points of orientation, but Hitler is not his new ‘Führer’. Also being the individualist he was, he found himself less inclined than ever to subordinate. With his sharp sense for the anti-intellectualist sentiments of the national-socialists he warns his generation of the ‘malignant aversion towards reasoning’. He starts to turn to the work and examples of certain intellectuals for guidance: people who reason for themselves and encourage discussion and thinking for oneself in others; not ‘Führer’. The people he names and cites form a rather heterogeneous mixture: the count Coudenhove-Calergi, the founder and foreman of the panEuropean movement; the philosopher Max Scheler; Ernst Bloch and Heinrich Mann. In 1927 the twenty-one year old Klaus Mann is an eclectic. He extracts whatever he can use from the work of this curious foursome to further his purposes, his reorientation. Without completely leaving behind his basic aesthetic attitude, he speaks of his new discoveries as ‘the other side’ and by doing so refers to “our social obligation” (ibid., p. 40) as pacifist Europeans. “And so by pursuing our lonesome adventure we have forgotten and neglected the other side. We are members of a threatened community, we are Europeans - woe be to us, as we neglect our duty! Then our 118
fantasies will remain meaningless, our lively and desperate monologues become uninteresting when we fall short on this other side. The other side is our social duty as the intellectual heirs to Europe [...] If it be possible for one of us, in the case of a new European war, not to participate in the protest which must follow? Is it conceivable that so much as one of us would allow himself to get carried away by this madness?” (ibid., p. 40) A few years later he acknowledges and specifies his position as “socialist humanism” (1992, p. 118), as “the spirit of 1789” and “the spirit of the October revolution” (ibid.). In 1942 he volunteers for service in the American army. The former pacifist participates in the war against National Socialism. I hope the few sources and citations given here serve to illuminate at least one thing: Klaus Mann's development from an apolitical young man, dealing with “moral, religious and aesthetic questions” in a “somewhat playful” (ibid.) manner (Mann, 1969 (1938)) into an anti-National Socialist intellectual with a well-developed sense of responsibility – such a developmental path was not mapped out for him in advance, let alone determined. Many young Germans, among them pupils of the Odenwaldschule, his generation, took another road by not resisting National Socialism, but by becoming followers. Did these fatal developments simply pass over the young generation in question? Were the followers of Hitler no more than the will-less victims of their “Generationslagerung”, of their “Generationszusammenhang” and their “Generationseinheit”? (Mannheim, 1952, p. 541) In which way were they active participants? Klaus Mann answers these questions with a clear ‘no’ and by doing so recognised his responsibility and the possibility of relatively free ‘choice’ (Sartre). The many possible ways of ‘participating’ presuppose some kind of activity by the person in question. He refused to view his own development in terms of the fate of a generation. Instead he saw it as a process about which he had more and more to say as time passed. As a youth and young adult the ‘child of his times’ was courageous and autonomous enough to turn away from the catastrophic tendencies of his era. Capable of learning and self-criticism, he didn’t just passively undergo his past influences (determinations) passively, he actively modified and utilised them. This applies especially to the background and the mentality of the German youth movement, the temporary influence of which has been my subject in this article. Cutting loose from this movement, from the cult of youth and turning away from his own generation are decisive themes in Klaus Mann's emancipation. He left the Odenwaldschule, which he remembered fondly, on his own initiative. And still, the early stories and the ‘romantic’ piece Anja und Esther are satu119
rated with the feeling of belonging to ‘the youth’ and with scenes focussing on the generation gap. However, in 1930 he writes in an open letter to Stefan Zweig: “There is also a form of ‘total-understanding’, a willingness towards the youths which goes too far. Not everything youths do points towards the future. This I write and I am young myself'. A large proportion of people of my own age, or those who are even younger still, have chosen to use the élan they ought to have reserved for progression, for regression. This we can never approve of” (Mann, 1992 (1930, p. 7)). At the end of this letter he notes: “Those born in 1902 could say: ‘La guerre-ce sont nos parents’. Should those born in 1910 then say: ‘La guerre-ce sont nos frères’? In that case it would be high time to start feeling ashamed of being members of a generation whose craving for action, whose radicalism, changed for the worst and turned into its opposite in such a terrible way” (ibid., p. 9). So as we see: even before the end of the Weimar Republic Klaus Mann had broken with the identifications of his generation. He realises that many of his peers lean towards the conservative end of the political spectrum, not wanting anything to do with democracy, reason or Europe (Mann, 1992 (1930, p. 84ff)). Instead they let themselves be fascinated by National Socialism. Klaus Mann resists the temptation of deducing this growing fascination “from the need of new bonds” (ibid., p. 60). He attributes the irrational passion for National Socialism to ‘the deep and fatal attraction in which violence as such is glorified, in which brutality as a principle is preferred to reason. This is plain language. It was written, not with analytical pretensions but rather as a confession of his ideas. The author is no longer speaking on behalf of the youths, of his generation, but against them. This also implies leaving behind the generation gap which, as we have seen, the youth movement had brought into the centre of its self-conscious examinations. ‘La guerre-ce sont nos parents’ – No! The dividing line between national socialists and anti-national socialists, between protagonists, sympathisers, hangers-on and people willing to offer resistance, which Klaus Mann now starts to concentrate on, cuts through all generations. “The natural front – the elderly versus the young – has lost all actuality and meaning” (ibid., p. 94). In 1933 Klaus Mann has turned his back to the majority of his generation. He doesn’t listen to his peer, the national-socialist leader von Schirach, but to his almost seventy year-old uncle the republican Heinrich Mann. Did the educators, the youth movement and the ‘Reformpädagogik’ which were so important to him during that one and a half year period have a (positive) influence on this reorientation? 120
Considering my arguments up until now the opposite would seem more plausible. Irrationality, disorientation and the craving for a ‘Führer’, cult of the community and extreme individualism, fetishism of youth, and the generation conflict which was almost blown up into a philosophy of life, all of these were elements of a mentality which was widespread within the youth movement during the Weimar Republic. In this paper I cannot attempt to answer the question whether it is correct to label this mentality protonational socialist or pre-national socialist, or perhaps only revealing some ideological affinity to ‘Hitler National Socialism’. My inquiry concerns the conditions and grounds which played a role in the transformation Klaus Mann experienced from an apolitical aesthete (and in so far a member of the post war youth generation) to a consciously political intellectual. It is perhaps not possible to give an unequivocal answer to this question. However it does seem that neither the intense, erotic-religiously flavoured cohabitation with selected peers, nor the innovative education he received, could have possibly unambiguously taught Klaus Mann to develop political consciousness or to choose a ‘correct’ political standpoint. The kinds of youth movement and ‘Reformpädagogik’ he encountered at the two ‘Landerziehungsheime’ were not committed to social and political causes in a positive manner. On the contrary, they remained somewhat aloof from these matters. It was no coincidence that the boarding schools were literally isolated, faraway from the cities: in the 'pedagogical province', so to speak. The young people, at least Klaus Mann and his friends, were entangled in the conflicts of puberty at the time; they kept themselves occupied with their own religious, erotic and artistic needs. The adults, at least Paul Geheeb and his associates, were so much held captive by their own humanist ideals about education, their unlimited belief in the child’s powers of selfrealization and a one-sided ‘leaving them to grow’-attitude towards the pupils, that they hardly gave a thought to setting up a compulsory and systematic learning process, let alone an open and uncompromising political curriculum. I have no idea whether the -putsch in March 1922 and Hitler’s ‘march to the Feldhernhalle’ from the 8th to the 9th of November 1923 were noticed and discussed at the Odenwaldschule, though presumably the elders didn’t strive to allow politics to infiltrate the school. Why concern oneself with Kapp, Ludendorff, Hitler or the industrialfinancial capital when one adheres to the adage ‘become who you are’, when one has practical work, the ‘Luftbad’, family-system and Herder, Goethe, Schiller and Wilhelm von Humboldt as ‘patron saints’? For Geheeb himself these neo-classical authors were no monuments, but true examples. In 1933 121
he too emigrated, after having condemned Hitler’s new government before the teachers and pupils attending the ‘school council’. On this occasion he described Hitler’s government as a “band of criminals at the head of the state” (cf. Feidel-Mertz, 1983, p. 29). I presume that what Klaus Mann knew about German fascism he had not learnt at school or through the youth movement, but as a result of the liberalintellectual climate of his family home and from National Socialism itself. This was also the way Klaus Mann himself perceived it. In Kind dieser Zeit he describes himself, a boy who has just left school, the Odenwaldschule, and who as “a seventeen-year-old was more antisocial and politically indifferent than he used to be when he was thirteen and fourteen” (p. 180). He also wrote about the select group of bourgeois intellectuals who, like himself, showed indifference to politics at a younger age, while offering resistance later on; as a thirty-two-year-old emigrant he noted: ‘Fascism was the hard school for these young intellectuals’ (Mann, 1969 (1938, p. 65)). Klaus Mann learnt more from this school about, for and against ‘life’ than most of his generation-unit. One of the problems at least seems to have been, that the generationunits of the 1920s and beginning 1930s (with quite differing formative experiences) finally united and formed not only a horizontal unity but also one which connected the different age-groups. Seen from this perspective Mannheim’s statement that the “phenomenon of generations is one of the basic factors contributing to the genesis of the dynamic of historical development” (Mannheim, 2000 (1936), p. 320) is in need of differentiation.
References Andreesen, A. (1934). Gegenwartsaufgaben der Landerziehungsheime. Die Erziehung, IX, 353-362. Bartels, I. (2002). Zum inflationären Gebrauch des Begriffs ‘Generation’ im aktuellen Mediendiskurs. Vokus. Volkskundlich-kulturwissenschaftliche Schriften, 1, Universität Hamburg. Online edition. Bude, H. (2000). Die biographische Relevanz der Generation. In: M. Kohli & M. Szydlik (Eds.), Generationen in Familie und Gesellschaft (pp. 19-35). Opladen: Leske + Budrich. Edmunds, J. & Turner, B. S. (2002). Generations, culture and society. Buckingham: Open University Press.
122
Elder, G. H. (1999). Children of the great depression. Social change in life experience (25th anniversary edition). Boulder: Westview Press. (Original publication 1974) Feidel-Mertz, H. (Ed.) (1983). Schule im Exil. Die verdrängte Pädagogik nach 1933. Reinbek by Hamburg: Rowohlt. Huguenin, E. (1926). Die Odenwaldschule. Weimar: Böhlau. Kohli, M. (2003). Generationen in der Gesellschaft. In R. Schmidt (Ed.), Systemumbruch und Generationenwechsel. Mitteilungen (H. 9, 9-18). Jena: Sonderforschungsbereich 580. Kohli, M. & Szydlik, M. (Eds.) (2000). Generationen in Familie und Gesellschaft. Opladen: Leske + Budrich. Kurzweil, Z.W.(1973). Die Odenwaldschule (1910-1934). Paedagogica Historica, XIII, 23-56. Mann, K. (1925a). Anja und Esther. Ein romantisches Stück in sieben Bildern. Berlin: Österheld. Mann, K. (1925b). Die Jungen. In: K. Mann, Vor dem Leben. Erzählungen. Hamburg: Enoch. (Original work published 1923) Mann, K. (1925c). Vor dem Leben. Erzählungen. Hamburg: Enoch. Mann, K. (1965). Kind dieser Zeit. Mit einem Nachwort von W. L. Shirer. München: Nymphenburger Verlagshandlung. (Original work published (1932). Berlin: Transmare) Mann, K. (1969). Die Kriegs- und Nachkriegsgeneration. In: K. Mann. Heute und Morgen. Schriften zur Zeit (pp. 206-227). M. Gregor-Dellin (Ed.). München: Nymphenburger Verlagshandlung. (Original work published 1938) Mann, K. (1976). Der Wendepunkt. Ein Lebensbericht. München: Spangenberg/ Ellermann. (Original work published (1942). The turning point. New York: L. B. Fisher). Mann, K. (1987). Briefe und Antworten 1922-1949. M. Gregor-Dellin (Ed.). München: Ellermann Verlag. Mann, K. (1990). Maskenscherz. Die frühen Erzählungen. Herausgegeben und mit einem Nachwort von Uwe Naumann. Reinbek: Rowohlt. Mann, K. (1992). Die neuen Eltern. Aufsätze, Reden, Kritiken 1924-1933. U. Naumann, M. Töteberg (Eds.). Reinbek by Hamburg: Rowohlt. – (pp. 16-18). Die freie Schulgemeinde. (Original work published (1924, February 21). Acht-Uhr-Abendblatt, Berlin). – (pp. 60-71). Fragment von der Jugend. (Original work published (1926). Die neue deutsche Rundschau). – (pp. 97-99). 'Der Alte'. (Original work published (1925). In: K. Mann. Vor dem Leben. Hamburg: Enoch). – (pp. 131-152). Heute und Morgen. Zur Situation des jungen geistigen Europa. (Original work published (1927). Hamburg: Enoch). – (pp. 254-274). Die Jugend und Paneuropa. (Original work published (1930)).
123
– (pp. 304-317). Wie wollen wir unsere Zukunft? (Original work published (1930)). – (pp. 318-320). Jugend und Radikalismus. (Original work published (1930)). Mann, T. (1961). Briefe I: 1889-1936. E. Mann (Ed.). Frankfurt a.M.: S. Fischer. Mannheim, K. (2000). The problem of generations. In: K. Mannheim, Essays on the sociology of knowledge. Collected works Volume five (pp. 276-322). P. Kecskemeti (Ed.). London: Routledge. (Original work published 1936) Naumann, U. & Töteberg, M. (1992). Einleitung. In: K. Mann, Die neuen Eltern. Aufsätze, Reden, Kritiken 1924-1933. U. Naumann & M. Töteberg (Eds.). Reinbek by Hamburg: Rowohlt. Pilcher, J. (1994). Mannheim’s sociology of generations, an undervalued legacy. British Journal of Sociology, 2, 481-495. Pinder, W. (1926). Das Problem der Generation in der Kunstgeschichte Europas. Berlin: Frankfurter Verlagsanstalt. Rang, A. & Rang, B. (1984). Reformpedagogiek en Nationaal-Socialisme in Duitsland. Comenius, 5, 272-296. Schänzler, N. (1999). Klaus Mann. Eine Biographie. Frankfurt/New York: Campus Verlag. Scheibe, W. (1969). Die Reformpädagogische Bewegung 1900-1932. Weinheim: Beltz. Shirley, D. (1992). The politics of progressive education: The Odenwaldschule in Nazi-Germany. Cambridge, MA: Harvard University Press. Turmel, A. (2004). Towards a historical sociology of developmental thinking: The case of generations. Paedagogica Historica, 40, 1-21. Vincent, J. A. (2005). Understanding generations: political economy and culture in an aging society. British Journal of Sociology, 56, 579-599.
124
10 The constitution of the subject and the possibility of critique. Response to Brita Rang STEFAN RAMAEKERS
A week ago my wife was looking at some family pictures with our oldest son (he’s five). One of these pictures showed my brother with a former girlfriend of his. It has been two years since my brother broke up with her, and our son did not remember her, so he asked who she was. My wife told him who she was and explained the situation, in terms she thought to be appropriate for his age, saying things like ‘sometimes people stop liking each other’, and that when that happens they no longer stay together, and that they then sometimes find someone else they do like, etc. “Oh, they’re divorced”, he replied, with a tone of voice that expressed something like “why didn’t you say so and spare yourself the entire explanation”. (They did not “divorce”, of course, simply because they were not married, but that’s not the point here.) I could not help but think: children nowadays grow up in a different world to the one I grew up in. When I was five, ‘being divorced’ was not a concept and a reality I was familiar with. Only later did I get to know this word and the reality it expressed. I also recall that whenever the word was used it was accompanied by an atmosphere of reserve, secrecy, and sometimes even shame and sinfulness - the tone of voice was markedly different from the one used by our son. ‘Being divorced’ was not part of the way I was initiated into this world in the same way that it is for our oldest son. He feels differently about it than I do. And I do not mean this in a morally judgmental sense, but, in a more basic sense, that my initial affective reaction to it is different than his. For him, coming to know the concept of ’divorced’ is something as normal as coming to know the colour ’green’, or learning about the concept ’round’, or about the north pole. The immediate reasons for this are obvious: schools, e.g., simply had to deal with this reality, at least if they still wanted to celebrate Mother’s Day, or Father’s Day, or if they still wanted to organise the annual 125
school party; society in general, partly because of the ongoing secularization, has gradually become more tolerant of the phenomenon; etc. Our oldest son has picked up the concept of ’being divorced’ as a matter of course. In contrast, I got to know it with a special connotation - that was different from learning about the north pole. ’Feeling differently about something’ in a sense might be the pre-eminent mark in determining what it means to belong to different generations. The sense I have in mind is the one Brita Rang hints at, in the first section of her text, when quoting Klaus Mann about how strange, even incomprehensible, older and younger people’s respective ‘complex feelings’ (p. 107) seem to one another. The existence of such differences in feelings -‘Gefühlsdistanz’ one could say - should not surprise us. It is a natural consequence of the intuition that education is in some sense, particularly in its early stages, an initiation into particular practices. Being initiated into practices is not just about learning to use language, it is being led into the totality of “agreement of judgements” (Wittgenstein, 1958, #242) which make up the world for us, which “determine” what is true and false, beautiful and ugly, good and bad, appropriate and inappropriate, just and unjust, green or red, round or square, etc. The normativity, that is the constitutive force of these agreements should not be underestimated. Ludwig Wittgenstein speaks of this normativity as “the hardness of the soft” (Wittgenstein, 1961, p. 44). With this he conveys that what seems at the outset to be rather ’soft’, i.e. ’merely human’ agreements, viz. cultural and social accretions, linguistic valuations, something which we can (supposedly) oppose to the hard data of biology, of genetic destiny, is deeply constitutive - and in that sense it is ‘hard’ - of the way we see, understand, feel about the world. Wittgensteinian agreements are, we could say, embodied. As agreements they are not articulated; rather, they manifest themselves in what we say and do, in how we speak and act, in how we feel. Nietzsche’s way of putting this can be helpful: “ … behind feelings there stand judgments and evaluations which we inherit in the form of feelings (inclinations, aversions)” (Nietzsche, 1982, # 35). Being initiated into particular practices, as coming to entering into a totality of agreements in judgements, is acknowledging that valuations become a part of us in the shape of feelings. If we can accept this, then the difference in the kind of relationship one can have towards a previous generation respectively one’s own generation becomes clear. Establishing a distance from the (or a) previous generation is not so much of a problem - on the contrary, with regard to the previous generation the issue is probably rather one of finding the appropriate rap126
prochement because of the very gap of feelings, the Gefühlsdistanz. The real difficulty is to find the appropriate distance towards one’s own generation, or put differently, to learn to feel differently. It is this issue in Rang’s paper that aroused my interest. Rang states that Mann’s “image of a generation was formed but also problematised by school experiences and developments of the youth movement” (p. 107). However, it is not clear to me how Klaus Mann, in contrast to his contemporaries, did in effect manage to turn himself away from his generation’s “catastrophic tendencies” (p. 119), apart from Rang saying that he “recognised his responsibility and the possibility of relatively free ‘choice’” (p. 119), or that he recognised that participation also always “presupposes some kind of activity by the person in question” (Ibid.). To be sure, Rang is correct in bringing this important point to the fore. In Wittgensteinian terms: agreements in judgements are not only passively endowed, they are also always collectively upheld by and in a particular community of competent language speakers. In this sense, there is always an indispensable human contribution to these agreements. But this does not resolve the issue since it simply puts off the question of how Klaus Mann came to realize this responsibility, this active involvement of his participation in a generation. There is no easy answer to be expected to the question as to how to distance oneself one’s own embeddedness. I do not think I am overstating if I say that the matter of a subject’s constitution and the possibility of critique is one of the most fundamental educational issues. (Or at least I think it is an issue philosophers of education should be concerned with.) Allow me, within the limitations of this reply, to touch upon one small element here, in a very brief attempt to try to dispel a powerful picture that always seems to hold us captive in discussions like these. Whenever initiation is at stake, the reproach of conformism and even of conservatism is never far away, while at the same time a longing for newness, for redemption is expressed, or at least implied. With the concept of ’generation’ this does not seem to be otherwise, as Klaus Mann’s history testifies. Moreover, the concept of generation even seems to be quite suited for the discourse of those who cherish hopes for a refreshing renewal. This is nicely illustrated by a recent review, in a Flemish newspaper, of the newly released debut CD of a band called Arctic Monkeys. The reviewer explicitly lets the concept generation do its work: “However, their sound is not, after having carefully studied everything, the product of an older band, but the outcome of four adolescents who have been enthusiastically learning to play for three years. Because of that the band sounds first and foremost like itself 127
and does not seem to revel in nostalgia for the past. Arctic Monkeys finally opens the generation gap wide again, whereas other recent bands meet with approval mainly because they sound so much like bands from 20 or 30 years ago” (De Meester, 2006, p. 22). Clearly, the idea that the generation gap is to be celebrated – for it is our guarantee of renewal – is linked here with the familiar theme that renewal should come from the younger generation. But what I wish to draw attention to here is something else, viz. the reiteration of the theme of renewal as coming from something “pure”, something not perverted by (alien) constraints. Put differently, the picture holding us captive is that initiation is, in some literal sense, in-corporation, bringing a body (corpus) into something else. However, if we take education as initiation as discussed above seriously, i.e. as coming to feel in a particular way, or again, if we take Klaus Mann’s suggestion seriously, i.e. that generations differ in complex feelings (as quoted by Rang), then what happens in initiation (or in the formation of a generation) is not so much in-corporation as rather the process of receiving a body. Education as initiation (or the formation of a generation) is in a relevant sense a process of em-bodying, of bestowing a body upon someone. At first sight, this may seem to make things even worse than before - for conformism and conservatism now seem to be inherent in the concepts of initiation and generation. However, given the observation that human beings need to apply in the future what they have learned, that is, given the observation that human beings do not just copy what others initiate them into (learning through imitation, e.g., is more than copying another’s behaviour), it would be more correct to say that it is exactly change and renewal that is inherent to the concept of initiation. Stanley Cavell’s words can be helpful here: “If what can be said in a language is not everywhere determined by rules, nor its understanding anywhere secured through universals, and if there are always new contexts to be met, new needs, new relationships, new objects, new perceptions to be recorded and shared, then perhaps it is as true of a master of a language as of his apprentice that though ’in a sense’ we learn the meaning of words and what objects are, the learning is never over, and we keep finding new potencies in words and new ways in which objects are disclosed. The ‘routes of initiation’ are never closed.” (Cavell, 1979, p. 180) Another way of putting this would be to say that initiation in the form of passing on what is valuable for us, is also always trust in it being out of one’s hands. In any case, it should be clear that both the fear of conservatism and the fixation on some kind of pure nature (as our only means of redemp128
tion) are expressions of a failure to acknowledge the dynamics inherent in the concept of initiation, or alternatively, inherent in belonging to a generation. At the same time it needs to be acknowledged that the change envisioned here is like the change of a river-bed: slow, but unmistakable. And because of its slowness, it will probably not be tomorrow that the change becomes visible. As said, it has not become clear to me, on the basis of Rang’s paper, why one person does and another person does not succeed in freeing herself from her own generation, from the complexes of feelings that co-constitute ones’ generation. The beginning of an answer lies perhaps in what Rang says (p. 109) fascinated the fifteen and sixteen year old Klaus Mann, i.e.: “something the educators had not intended”. Perhaps finding the appropriate distance, turning oneself away from one’s generation, as Klaus Mann did, is also something which we need to allow to be out of our hands
References Cavell, S. (1979). The claim of reason. Wittgenstein, skepticism, morality and tragedy. Oxford: Oxford University Press. De Meester, K. (2006, January 20). De Gapende generatiekloof is terug. De Morgen, p. 22. Nietzsche, F. (1982). Daybreak. Thoughts on the prejudices of morality (R.J. Hollingdale, Trans.). Cambridge: Cambridge University Press. Wittgenstein, L. (1953). Philosophische Untersuchungen/Philosophical investigations (G. E. M. Anscombe, Trans.). Oxford: Basil Blackwell. Wittgenstein, L. (1961). Notebooks, 1914-1916 (G. E. M. Anscombe, Trans.). Oxford, Basil Blackwell.
129
11 The contingency of scholarly development across generations FRIEDA HEYTING
In the world Miguel de Cervantes created for his Don Quixote four hundred years ago, romances of chivalry were very popular. Yet in extracting his destiny from these romances, Don Quixote was unique amongst his fictitious contemporaries. If I rely on my own experience, this unpredictability of what learning will result from either reading books or attending lectures still applies to modern academic life. University teachers invest a lot of energy in determining exactly what knowledge and skills they wish their students to acquire. Fortunately, they often succeed in realising such aims to a large extent. In the end, most students are aware of the major theories in their field and are able to execute elementary research techniques. However, this model of knowledge can at best partly explain how scholarly knowledge is passed on and further developed from one generation to the next. Again judging from my own experience, the most lasting and decisive influences appear to be exerted by insights that occurred to me whilst reading or listening to subject matter dished up for entirely different purposes. Reconstructing a few of the most striking episodes from the history of my own developing insight chiefly substantiates the contingency of learning, teaching, and scholarly development in general. In the early sixties, one of my exams in social psychology was about the development of prejudice. Compulsory reading included Gordon Allport’s (1958) The Nature of Prejudice and some related articles. It made an indelible impression on me, but only partly due to its subject matter, prejudice, which was such an all-pervasive issue in those days. Reading that young children fail to notice the different skin colours of their peers, it suddenly occurred to me that people do not simply see whatever they lay their eyes on; they learn what to see. Writing about the non-natural nature of emotions, Martha Nussbaum (2004, p. 35) recently made a similar observation: children hardly notice skin colour as a salient fact about people, unless it is made 131
salient to them by adults. To me, this information evoked the baffling insight that I might have to reckon with a yawning gap between observed realities on the one hand, and reality itself on the other. It was this insight, which first put me on the track of philosophy. However, this newly acquired insight did not make my life easier. By making reality itself inaccessible to me, well hidden behind the masks of its observers, it rather seemed to force me into the uncomfortable position of the knight errant Don Quixote, who tries to convince his squire Sancho Panza not to depend on his eyes. The windmills in the distance might well disguise the enemy forces his knightly honour obliges him to defeat, because, Don Quixote explains: “Matters of war, more than any others, are subject to continual change” (Cervantes, 2005, p. 59), and “all things having to do with knights errant appear to be chimerical, foolish, senseless, and turned inside out” (ibid., p. 195). While pondering this, I kept wondering about the gap between observation and reality. It took a second flash of insight before I could formulate a first cautious conjecture about its nature. It came to me while reading Jean Paul Richter’s (1975) Levana oder Erziehlehre (“Levana or the theory of education”), written in 1806. It talks about a professor of education, who erroneously meant to do his benefactors a favour by using his inaugural lecture to prove the inconsequential nature of upbringing, no matter whether administered at school or at home. How could the influence of a teacher or a parent, Jean Paul’s professor argues, exceed that of the uncountable people, opinions, and actions that are all part of what might be called the ‘Zeitgeist’? (p. 538) Moreover, he continues, those individual teachers and parents themselves became what they are because of this same Zeitgeist. We can consider ourselves fortunate, he concludes, because of the vainness of all educational effort. Imagine what the world would look like if it were otherwise: it would be populated with clones of well-meaning educators! (p. 544). Jean Paul’s protagonist, however, was less fortunate. He had misjudged his audience. The public did not feel relaxed by the idea of occupying inconsequential jobs from which no harm was to be expected. If anything, Jean Paul’s public was extremely offended. No sooner had our professor finished his inaugural lecture than he was dismissed, and no sooner was he dismissed than he announced his valedictory lecture, which dealt with the decisive influence of upbringing on child development. Interpreting his inaugural lecture as an example of sophism, an exercise in idle philosophical argumentation, he now got ready to prove how strong an influence upbringing really 132
exerts on child development (p. 549). As the professor points out, the spirit of the age and the people as a whole do not exert any influence as such, but only through the devotion of an individual friend, a teacher, or a family – who undoubtedly do make the most profound impression during childhood (554). From a distance, the masses may seem all seem alike, but a closer look will reveal immense individual differences. Observing the globe as a whole may hide any mountain from view, as observing the mountain may leave the stony path across invisible, but walking the path will make each stone on its surface tangible, as the professor concludes (p. 555). In fact, I drew more than one lesson from this story. By demonstrating how observation could support quite opposing versions, it confirmed my suspicions that it would be unwise for a scholar solely to depend on his eyes if he wants to get a grip of educational reality. But, and here is the first lesson I drew from Jean Paul’s story, it also suggests that reality does not appear to us in an arbitrary way. Don Quixote did not see an arbitrary world; whatever he saw, he saw as a knight, always prepared for the possibility of having to fight whatever might threaten the safety or sovereignty of his king. And whatever the audience of our professor heard, they heard so in their capacity as professional educators, always prepared for the possibility of having to correct whatever might lead a developing child astray. Apparently, the world is in the eye of its observer, exposing itself in a functional way, ready to be acted upon. This idea of the instrumental nature of observation gives a first glimpse of what the gap between observation and reality might be like. However, it does not ‘fill’ the gap, and it does not enable us to reconstruct reality as it ‘really’ is after all. On the contrary, it only reinforces the scepticism Allports book about prejudice had given rise to, and it also reinforces the Quixotean discomfort of always having to doubt our observations. Despite the almost fixed rigidity of his knightly perspective, Don Quixote could never be certain whether it would yield the results he was aiming at. Though it was clear he could not rely on appearances because a knightly world always runs the risk of being distorted by the Wise Frestón or another enchanter, he could never say with certainty whether, and if so, to what effect they might have distorted the outward appearances of his world. Against this background, the philosopher Steven Nadler (1997) even compares Cervantes’s Wise Frestón – who always seems prepared to enchant Don Quixotes world in the most devious of ways – with the ‘malicious demon’ as introduced by the early modern philosopher Descartes. As Descartes’s demon was meant to under133
mine unfounded faith in reason by tirelessly raising objections, Cervantes’s enchanters were meant to undermine implicit faith in direct observation. Jean Paul’s story leads to the same conclusion. It demonstrates that educators look at childhood from a very specific perspective – a perspective that is dominated by the idea of ‘influence’. At the same time, however, this story demonstrates the almost complete inability of educational scholars at that time, including Jean Paul himself, to understand the nature and workings of this educational influence in any theoretically founded way. This resulted in the second lesson I drew from the short career of Jean Paul’s professor. Coming from a background in the natural sciences, it struck me that this problem of lacking educational theory was far from solved when I first read the story. It still isn’t today. Of course I am well aware, that educational theories are now far more elaborated, more refined, and better tested than when Jean Paul wrote his story two hundred years ago. However, that still leaves much to be desired. For example, whereas natural science can explain in minute detail how and why a bolt of lightning striking a tree is processed, resulting in a carbonised stake, learning theories – though capable of casting an illuminating light on a lot of developmental phenomena – still lack the potential to keep a close track of the chain of events that transform a teacher's lesson into a specific capability in the pupil. I do not mean to hold up physics as an example for educational scientists, nor am I accusing them of being blind to theoretical shortcomings. It might well be the case, that pushing back frontiers in educational explanation has to reckon with almost incalculable complexity, and unlike Don Quixote, who never even felt tempted to adapt his knightly theories to the sorry incidents that proved them wrong, educational journals and conferences buzz with theoretical considerations, revisions and controversies. And, what’s more important, any science, including physics, will meet the limits of its theoretical capabilities at some point. But the realisation of the defective nature of educational theory did push me towards philosophy once more, after the gap between observation and reality had first instigated it. What exactly are educational explanations, and what makes them convincing? These questions and related ones appeared to occupy the students of a whole sub-discipline: philosophy of education, in those days – about forty years ago – still often revealingly called ‘educational theory’. In spite of this suggestive name, its practitioners were not engaged in creating the theories their empirical colleagues had as yet failed to produce. And indeed, what would have entitled them to do so, having no privileged access to edu134
cational reality either? The name ‘educational theory’ was inspired by the theoretical state of the discipline in quite another way. Poor explanations usually, and often implicitly, derive compensatory sustainment from seemingly obvious lines of thought and from common sense. Because of that, philosophers of education concentrated – and still concentrate – on the adequacy, consistency, and interrelatedness of educational argumentations and concept use. By focussing on argumentation, logic, and concept-use, they are sometimes able to unearth those inconsistencies or incompatibilities that result from the typical mix of convention and theoretical refinement that is characteristic of educational explanations. For example, Michael Luntley’s (2005) recent analysis of learning conceptions suggests that explanations of teaching and learning might especially gain from a sharper distinction between learning as getting familiar with accepted theories and knowledge – which he calls ‘mimicry’ (p. 703) – and learning as developing capacities for recognising and shaping the conceptual patterns and the reasons that figure in justification. However, this mainly analytical and critical function did not reflect what philosophers of education appeared to consider – and still consider – their ‘core business’. Even today, philosophers of education almost invariably motivate the definition of research problems by pointing out their practical relevance, and they present the results in the same light. Luntley’s (2005) analysis, for example, concludes with demonstrating how considerations about what kind of curriculum should be considered ‘best’ could benefit from the results of his analysis. It is this concern for the quality of educational practice that seems most characteristic of philosophy of education. This ethical or – more generally – axiological concern still outshines its conceptual and analytical orientation by far. In itself, that is not surprising, considering that the whole educational discipline seems to function as the scientific counterpart of a major field of practice in society (cf. Luhmann, 2002). However, at first it did not seem clear to me at all what kind of pretensions philosophy of education would be able to live up to with respect to this primarily axiological concern. Could philosophy of education claim to be the legitimate arbiter in issues of value? And if not, what kind of value-related claims could it substantiate? With respect to these issues, my then teacher, Helena Stellwag, soon – unintentionally – opened my eyes for me. In the sixties and early seventies of last century, Stellwag distanced herself more and more from any prescriptive ambitions in philosophy of education. In those days such ambitions mainly concentrated on deriving educa135
tional directives from well-defined ideological systems, which resulted in a wide array of educational doctrines. According to Stellwag, philosophers of education should temper their ambitions, and restrict themselves to mutually comparing those ideological systems and their derived educational prescriptions, and to working out whether they met conceptual and logical standards (see Stellwag, 1962). Once, confronted with the then almost paradigmatic question as to whether philosophy could also assess the moral quality of those educational doctrines, she answered: “Fortunately we are not in the position to do so. Imagine how terrible it would be if philosophy could ever give a decisive answer about what ‘the good’ might entail – it would rob us of our very humanity”. Despite its paradoxical import, this reply brought an important insight home to me. It was triggered by the expression ‘the good’ – Stellwag was a classicist –, which made me realise that ‘valuableness’ was often treated as if it could exist – separate from a valuator, waiting to be found, presumably by an expert-philosopher. That was the way Don Quixote had seen it, taking the knightly doctrine he had extracted from his romances of chivalry for an objective fact and refusing Sancho Panze any attempt to discuss it. Realising the unfoundedness of this approach, I suddenly felt confronted with an even wider gap than that between observation and reality. However inaccessible reality did seem to me, it still seemed relatively safe to presuppose its existence. The ‘good’, as I now saw it, could not even be conceived as existing in its own right. It should be considered as ‘made’ rather than found, which principally made it a matter of discussion, and not of establishment. This insight had far-reaching consequences. To me, it did not only seem unthinkable to conceive of ‘the good’ as an externally existing object, the same had to apply to all mental entities, varying from the meaning of concepts to ideological systems and their derived educational directives. As a consequence, whole areas of philosophical investigation became futile to me. With respect to linguistic analysis, philosophy would have to refrain from any claims about the ‘correct’ meaning of a word. Besides I started wondering why one should investigate the moral merits or educational consequences of ideological systems that I no longer believed existed as such. In my view, the summaries of such systems – such as liberalism, or communism – could only be understood either as one specific version by one specific author or as an attempt to summarise what an indistinct group of people had maintained thus far. Both interpretations forbade any generalised – let alone prescriptive – claims about ideological educational thinking. More136
over, every practician could do nothing but make his or her own unique version, including preferred educational ends and approaches. I could think of no compelling argument for philosophy to anticipate any such choice, rather than leaving it to the persons concerned. In my view, all this consigned philosophy of education to its critical and analytical functions, back to the fringes of educational science, where it could best play its indispensable role. In his inaugural lecture, delivered in 1969, the philosopher Theo de Boer summarised this role as follows: “the philosopher ceaselessly tries, at the side of the road where science goes its own steady way, to alert to those dimensions that tend to be forgotten, precisely because they are constantly presupposed” (De Boer, 1969, p. 34, trans. FH). The philosopher is someone with a special interest in whatever slips through the nets scientists cast; as such he is actually someone who principally leads a marginal existence (idem, 1969, p. 30). I would have happily persisted in this marginal existence, if my last teacher and predecessor, Adalbert Rang, had not disturbed this relative peace. In one of our many discussions, he epitomised my position as ‘scepticism engagée’. It was only at that moment, that its paradoxical character started to fully get through to me, a process in which Rang’s valedictory lecture proved to be of much help. Discussing a similar paradoxy with respect to pluralism – of which he is an ardent advocate –, he explained that pluralism would not make sense as an absolute prerogative of diversity over unity. In his view, it would only make sense as an attitude towards balancing both extremes, never taking any division nor any unity for granted, but putting both up for discussion – not only among philosophers, but ultimately preferably in society itself (cf. Rang, 1994). Gradually I started developing from a ‘dogmatic’ sceptic into a ‘doubting’ sceptic, a process that will give me no rest and that is still far from concluded (cf. Heyting & Mulder, 1999). Cervantes (2005, p. 934 ff.) apologised at the end of his book for having created Don Quixote, thus luring other authors into imitation. In his turn, Don Quixote made apologies at the end of his life for having read those romances of chivalry and for imposing their contents on Sancho Panza. However, at the end of my career I do not think I will have to go as far as making apologies for reading, explaining, and writing philosophy. Inspired by Don Quixote’s empirical scepticism, but doubting his rationalist dogmatism – as well as the mimicry conception of teaching and learning it implies – I think I can safely consign my work to the next generation, because I am sure they will decide for themselves what to do with it. 137
References Allport, G. W. (1958). The nature of prejudice. New York: Doubleday Anchor. Cervantes, M. d. (2005). Don Quixote (E. Grossman, Trans.). London: Vintage. De Boer, T. (1969). Wijsgerige en wetenschappelijke antropologie. Assen: Van Gorcum & Comp. Heyting, G. F., & Mulder, E. (1999). Educating the sceptic: Sextus Empiricus and education. Paedagogica Historica, 35(2), 359-378. Luhmann, N. (2002). Das Erziehungssystem der Gesellschaft. Frankfurt a.M.: Suhrkamp. Luntley, M. (2005). The character of learning. Educational Philosophy and Theory, 37(5), 689-704. Nadler, S. (1997). Descartes's demon and the madness of Don Quixote. Journal of the History of Ideas, 58(1), 41-56. Nussbaum, M. (2004). Hiding from humanity. Disgust, shame, and the law. Princeton: Princeton University Press. Rang, A. (1994). Pädagogik und Pluralismus. In G. F. Heyting & H. E. Tentorth (Eds.), Pädagogik und Pluralismus (pp. 23-50). Weinheim: Deutscher Studien Verlag. Richter, J. P. F. (1975). Levana oder Erziehlehre. In N. Miller (Ed.), Jean Paul. Werke in zwölf Bänden (Vol. 9, pp. 515-639). München: Carl Hanser Verlag. Stellwag, H. W. F. (1962). Positief of negatief. Pedagogische Studiën, 39(7/8), 321332.
138
List of contributors
Frieda Heyting holds a chair in the philosophy and history of education at the Universiteit van Amsterdam, The Netherlands. Ernst Mulder is a teacher and researcher in the history of education in the Department of Education at the Universiteit van Amsterdam, The Netherlands. Stefan Ramaekers is a teacher and researcher in the philosophy of education in the Faculty of Psychology and Educational Sciences at the Katholieke Universiteit Leuven, Belgium. Brita Rang holds a chair in the history of education at the Johann Wolfgang Goethe-Universität in Frankfurt am Main, Germany. Doret de Ruyter holds a chair in the philosophy and history of education at the Vrije Universiteit in Amsterdam, The Netherlands. Leonie le Sage is a teacher and researcher in the philosophy of education in the Faculty of Psychology and Education at the Vrije Universiteit in Amsterdam, The Netherlands. Judith Suissa is a teacher and researcher in the philosophy of education at the Institute of Education of the University of London, United Kingdom. Christiane Thompson is a teacher and a researcher in the philosophy of education in the Department of Education at the Martin Luther Universität in Halle-Wittenburg, Germany. Roel van Goor is a teacher and doctoral candidate in the philosophy of education in the Department of Education at the Universiteit van Amsterdam, The Netherlands. Raf Vanderstraeten holds a temporary chair in sociology and sociological theory in the Faculty of Sociology at Universität Bielefeld, Germany, and is affiliated to the Faculty of Psychology and Educational Sciences at the Katholieke Universiteit Leuven, Belgium. Christopher Winch is a teacher and researcher in the philosophy of education in the Department of Education and Professional Studies at King’s College London, United Kingdom.
139