* Bourdieu, Pierre et Passeron, Jean-Claude, La Reproduction (Paris, Ed. de Minuit, 1970)Google Scholar, English translation: Reproduction (London and Beverly Hills, Sage, 1977)Google Scholar. Throughout references to Bourdieu and Passeron's Reproduction are taken from the English translation followed by the corresponding reference to the French original in brackets; Bernstein, Basil, Classes, Codes and Control (London, Routledge, 3 vols: 1971–1975)CrossRefGoogle Scholar.

(1) Karabel, J. and Halsey, A. H., Power and Ideology in Education (New York, 1977), p. 62Google Scholar.

(2) Archer, Margaret S., Social Origins of Educational Systems (London and Beverly Hills 1979), p. 54Google Scholar.

(3) Bernstein, Basil, Class, Codes and Control, Vol. 3, op. cit. p. 3Google Scholar.

(4) Bernstein, , Ritual in Education, Class, Codes and Control, Vol. 3, op. cit. p. 65–6Google Scholar.

(5) Bernstein, , A Critique of the concept of compensatory education, Class… Vol. 1, p. 200Google Scholar. Thus I would agree with Karabel and Halsey that ‘one of the main unresolved problems of Bernstein's, work [is] how it is that “power relationships penetrate the organization, distribution and evaluation of knowledge”, Power and Ideology in Education, op. cit. p. 71Google Scholar.

(6) The following statements from Bern-stein illustrate the axiomatic status accorded to direct penetration: Social class shapes ‘both the organizational structure and contents of education’, Postscript, Class, Vol. 1, p. 241; ‘The value system of the middle class penetrates the texture of the very learning context itself’, Social Class, language and socialization, ibid, p. 186; ‘Changes in the distribution of power and the principles of social control affect the what, how, where, when, and with whom, of school learning’, Postscript, ibid. p. 239; ‘Class relationships regulate the transmission, participation in and the possibility of changing dominant cultural categories’, Introduction, Class… Vol. 3, p. 23Google Scholar; ‘A school is a microcosm of society’, Sources of consensus and disaffection in education, ibid. p. 49.

(7) Bernstein, , Introduction, Class…, Vol. 1, p. 17Google Scholar.

(8) Bernstein, , Sources of consensus and disaffection in education, op. cit. p. 49Google Scholar.

(9) Compare the diagrams, Figures 1.1. and 1.2. Class, Codes and Control, Vol. 3, p. 21 and p. 24.

(10) The following figure summarizes these points by critically annotating one of Bernstein's basic diagrams. (Introduction, Class… Vol. 3, p. 21).

(11) Bernstein, , Introduction, Class… Vol. 3, p. 25Google Scholar.

(12) Bourdieu prefers the term ‘pedagogic action’ but pedagogic does not carry the more restricted connotations of this term in English.

(13) Bourdieu, Pierre and Passehon, Jean-Claude, Reproduction, op. cit. p. 5 (Fr. ed. p. 20)Google Scholar.

(14) Bernstein is explicit about this procedure when explaining why, in his theorizing, ‘the school is abstracted from its wider constraints’. He continues: ‘I think it is possibly because I am sensitive to inter-actions and once these are part of my experience, I can begin to intuit what I take to be the structural principles which they embody. But it takes a long time, at least for me, to condense experience into a concept, and even longer to extract from the concept what one has not in the first place directly experienced. Yet as we are all part of society, that experience is in us. When a set of concepts do not work, it is as if one's own sense of the social is obscured. When they do, then one's own biography is continuous with the formulation. Introduction, Class… Vol. 3, p. 2.

(15) ‘The relative strengths of classifications and frames determined for me the regulative principle of the transmission at a most general level’, ibid. p. 8.

(16) ‘The logical construct of a power relation manifesting itself nakedly has no more sociological existence than does the logical construct of meanings that are only cultural arbitrariness […]. There is no PA which does not inculcate some meanings not deducible from a universal principle (logical reason or biological nature): authority plays a part in all pedagogy, even when the most universal meanings (science or technology) are to be inculcated […]. It follows that PA, always objectively situated between the two unattainable poles of pure force and pure reason, has the more need to resort to direct means of constraint the less the meanings it imposes impose themselves by their own force, i.e. by the force of biological nature or logical reason’. Bourdieu, and Passeron, , Reproduction, p. 9–10 (Fr. ed. p. 24)Google Scholar.

(17) Ibid. p. 6 (Fr. ed. p. 19).

(18) Bourdieu's criticism of Weber, that ‘he overestimated the autonomy of the technical functions of both the educational system and the bureaucratic system relative to their social functions’ could be turned against him for indulging in a much greater exaggeration in the opposite direction. Ibid. p. 167 (Fr. ed. p. 205).

(19) Ibid. p. 7 (Fr. ed. pp. 21–2).

(20) The concept of an educational system, as used here, has no significance for Bourdieu who uses the term unconcernedly in traditional contexts, referring for example to ‘a system of education in which P[edagogic] A[ction] is not set up as a specific practice but falls to virtually all the educated members of a group or class’. Ibid. p. 64 (Fr. ed. p. 81).

(21) Ibid. p. 190 (Fr. ed. p. 225).

(22) Thus he quotes both Durkheim and Weber approvingly: ‘The former being justified in regarding the medieval university as the first true E[ducational] S[ystem] in the West’, the latter for considering ‘that the determining features of the educational institution are present with the appearance of a corps of permanent specialists’. Ibid. p. 56 (Fr. ed. pp. 71–2). These are merely aspects of the same ‘autonomization of a practice’, i.e. facets of progressive institutionalization.

(23) Ibid. p. 54 (Fr. ed. p. 70).

(24) Ibid. p. 55 (Fr. ed. p. 71).

(25) Archer, Social Orgins of Educational Systems, op. cit.

(26) Thus Bourdieu considers the decenand tralization of the British system to be unimportant compared with the force of the logic when he writes: ‘If one forgets all that the educational system owes to its essential function of inculcation, one fails to recognise the specifically pedagogic foundations and functions of standardizanent tion of the message and of the instruments of its transmission (a pedagogic homogenization which is to be found in the administratively most decentralized systems, such as, for example, the British system)’. Ibid. p. 171, n. 17 (Fr. ed. p. 182, n. 17).

(27) Ibid. p. 21 (Fr. ed. p. 36). This traditionalism may appear to imply that education has enough autonomy to be traditionalistic, were it not that traditionalism emanates from the dominant classes, which in France do not ask education to do ‘anything other than reproduce the legitimate culture as it stands’. Ibid. p. 59 (Fr. ed. p. 75).

(28) Instead, questions of causality (i.e. what accounts for the parallels detected) are displaced to a higher level—outside the sociology of education altogether and significantly they become even more ‘culturalist’. Thus Bernstein resorts to the induction of more general regulative principles (ideational in nature) which generate the parallelism and enter the realms of speculative philosophy and anthropology. Bourdieu on the other hand proceeds deductively, seeking to account for the homologies he perceives between church and school or school and state by appeal to a supposed theoretical unity underlying all actions of ‘symbolic violence’—a claim which depends inter alia on the possibility of ignoring structural variations not only within education but between social institutions as diverse as religion, government and education.

(29) Bourdieu, and Passeron, , op. cit. p. 7 (Fr. ed. pp. 21–2)Google Scholar.

(30) Ibid. p. 38 (Fr. ed. p. 53). Berntein, on the other hand, deals mainly with the subjective conditions for misrecognition, namely that the distance between the fundamental division of labour and the ultimate endorsement of a philosophy of education is sufficiently great to conceal the dependence of the latter on the former (concealment being a prime condition of maintenance), whilst the philosophy appears to confer independent cultural approval on present class-related practices.

(31) Ibid. p. 67 (Fr. ed. p. 83).

(32) Ibid. p. 15 (Fr. ed. p. 29).

(33) Bernstein, , Class and Pedagogies: visible and invisible, Class… Vol. 3, p. 123Google Scholar.

(34) Thus the new middle class can press onwards with their progressive pedagogy: ‘And this can be done with confidence for the secondary school is likely to provide both visible and invisible pedagogies. The former for the middle class and the latter for the working class’. Idem.

(35) ‘We have a homologue between the interruption of the new middle class of the reproduction of the old and the interruption of the new educational pedagogy of the old; between the conflict within the middle class and the conflict between the two pedagogies’. Ibid. pp. 122–3.

(36) Thus instead of leading to struggle and a weakening of domination, the reproduction of the dominated cultural arbitrary operates so as to reinforce and reproduce the power differentials which distinguished between the two cultures in the first place. For simultaneously the dominant PA will successfully have reproduced the intellectual and moral integration of the dominant group or class.

(37) Bourdieu, and Passeron, , op. cit. p. 167 (Fr. ed. p. 205)Google Scholar.

(38) Bernstein, , On the curriculum, Class… Vol. 3, p. 81Google Scholar.

(39) See Archer, , Social Origins of Educational Systems, op. cit. for a discussion of this theme in four countries: Denmark, pp. 449–463Google Scholar; England, pp. 516–568; France, pp. 306–309, 321–323, 327–331, 344–355, 357–359; Russia, pp. 285–289, 302–305.

(40) ‘Delegation of the right of symbolic violence […] entails the impossibility for that agency of freely defining the mode of imposition, the content imposed and the public on which it imposes it (the principle of the limited autonomy of pedagogic agencies) Reproduction, p. 27 (Fr. ed. p. 42).

(41) Both relative autonomy and the reason for it are juxtaposed in the following quotation: ‘Would the freedom the system allows to the agents appointed to inculcate be so great if it were not conceded in return for the class functions which the university never ceases to fulfil’. Ibid. p. 114 (Fr. ed. p. 143).

(42) Ibid. p. 32 (Fr. ed. p. 47).

(43) Ibid. p. 60 (Fr. ed. p. 76).

(44) The last proposition, 4·3, states that as far as the dominant educational system is concerned, then ‘by means proper to the institution, it produces and reproduces the necessary conditions for the exercise of its internal function of inculcating, which are at the same time the sufficient conditions for the fulfilment of its external function of reproducing the legitimate culture and for its correlative contribution towards reproducing the power relations’ (my italics). Ibid. p. 67 (Fr. ed. p. 83).

(45) Bernstein, , On the classification and framing of educational knowledge, op. cit. p. 94Google Scholar.

(46) Bernstein, , Open Schools open society?, Class… Vol. 3, p. 69Google Scholar.

(47) Bernstein, Introduction, ibid, p. 7.

(48) Bernstein disallows any importance being attached to his illustrations when he writes that ‘the descriptive statements have been selectively patterned according to their significance for the theory’. This is intended to reduce the former to a subordinate or auxiliary role and implies a clear separation between descriptive and theoretical statements. Apud On the classification and framing of educational knowledge, op. cit. p. 95.

(49) ‘Book I is the product of a long series of transformations, all tending to replace existing propositions with other more powerful ones which in turn generated new propositions linked to the principles by more and closer relations’, ibid., p. ix–x (Fr. ed. p. 10).

(50) Reproduction, p. ix (Fr. ed. p. 9).

(51) Archer, , Social Origins of Educational Systems, p. 173–183 ffGoogle Scholar.

(52) Bernstein, , Class and Pedagogies: visible and invisible, op. cit., p. 120–1Google Scholar.

(53) Yet when later reflecting on this paper Bernstein fully accepts that this kind of private education is far from being universal: ‘I was very impressed with the spectrum of British public schools, which over the last hundred yeats has created a range of social types out of the beatings of Harrow and the subtle sponteneity of Summerhill […] I know of no other middle class which has the possibility of such a differentiated form of socialization’. Introduction, Class… Vol. 3, p. 18.

(54) Bernstein, , Open Schools—open society? op. cit. p. 69Google Scholar.

(55) Bernstein, , on the classification and framing of educational knowledge, op. cit. p. 100Google Scholar.

(56) Reproduction, p. 28 (Fr. ed. pp. 43–4).

(57) Ibid. p. 60 (Fr. ed. p. 76).

(58) Ibid. p. 58 (Fr. ed. p. 74).

(59) Ibid (ibid.).

(60) For details of these measures see Vaughan, M. and Archer, M. S., Social Conflict and Educational Change in England and France: 1789–1848 (Cambridge 1971), chs. 7 and 8Google Scholar.

(61) For example in England, voluntary training colleges (mainly run by religious denominations) outnumbered those run by local educational authorities between 1902 and the Second World War. Today public colleges, though now in the majority, still do not monopolize teacher training, the ratio being just over 2:1. See Taylor, William, Society and the Education, of Teachers (London 1969)Google Scholar.

(62) Reproduction, p. 57 (Fr. ed. p. 72).

(63) Ibid. p. 55 (Fr. ed. p. 71).

(64) This rudimentary classification scheme is fundamentally skewed since characteristics shared only by centralized systems constitute one pole of the supposed continuum which runs between ‘(1) a system of education in which P[edagogic] A[ction] is not set up as a specific practice but falls to virtually all the educated members of a group or class (with only sporadic or partial specialization), and (2) an E[ducation S[ystem] in which the P[edagogic] Au[thority] necessary for the exercise of P[edagogic] A[ction] is explicity delegated to and juridically reserved for a corps of specialists, specifically recruited, trained and mandated to carry out P[edagogic] W[ork] in accordance with procedures controlled and regulated by the institution, at fixed times and in fixed places, using standardized controlled instruments’. Ibid. pp. 64–5 (Fr. ed. p. 81).

(65) Archer, , Social Origins of Educational Systems, op. cit. See p. 173–182 ffGoogle Scholar.

(66) See note 14.

(67) Bernstein, , Ritual in Education, op. cit. pp. 58–60Google Scholar.

(68) Ibid. p. 63.

(69) Reproduction, p. 59 (Fr. ed. p. 75). Historically this is simply not the case: from the July Monarchy onwards the system has been bombarded with demands for modernization, for technical specialization and for instrumental relevance— through the Second Empire, the Third, Fourth and Fifth Republics. Certainly these have only tardily won inadequate concessions, but the point is to understand why they have been systematically frustrated rather than denying the existence of such demands.

(70) In turn this cultural continuity itself is accounted for by the fact that academic traditionalism ‘despite the change in the social structure, has always managed to occupy homologous positions in the system of relations which link it to the dominant classes’. Again this leads full circle back to the problems of unexplained correspondences. Reproduction, p. 129–30 (Fr. ed. p. 165).

(71) As for example in the following quotation: ‘The Revolution and Empire continued and completed a tendency which had already begun under the Monarchy. Besides the Concours Général, which, set up in the 18th century, extended on to a national scale the competition taking place in each Jesuit college and consecrated the humanistic ideal of belles-lettres, the agrégation, re-established by decree in 1808, was first organized in 1776, in a form and with a significance very close to those it has today. If such facts, and more generally everything relating to the education system's own history, are almost always ignored, this is because they would belie the common representation which […] would have it that the French system owes its most significant characteristics to Napoleonic centralization’. Ibid. p. 171, n. 17 (Fr. ed. p. 182, n. 17).

(72) Bernstein, , On the classification and framing of educational knowledge, op. cit. p. 90Google Scholar.

(73) Bernstein, , On the curriculum, Class… Vol. 3, p. 82Google Scholar.

(74) As it stands, the reproduction thesis gives an excellent account of certain sociocultural processes which take place once a system has been structured in a particular way, but provides an inadequate explanation of this structuration. To get full value from Bourdieu's theory, one needs to know when (historically) and where (comparatively) it works most and least well. Otherwise it remains a kind of ‘extreme’ type whose approximation to empirical reality varies greatly for different countries and different times. This in turn needs explanation, but as it stands the theory cannot provide it because of its ahistorical and noncomparative nature.

(75) Thus, for example, both of Bernstein's key concepts, ‘classification’ and ‘framing’, upon which the typology of ‘knowledge codes’ is built, seem to bear a relationship to the structure of educational systems which is well worth exploring. Strong and weak framing appear to coincide respectively with the centralized and decentralized decentralized system, whilst the un-specialized and specialized ‘collection codes’ also seem to parallel this systematic division. In other words, different kinds of educaional systems could be seen to furnish the conditions and controls necessary to initiate and sustain different types of curricular, pedagogical and evaluative practices and this could be established by the comparative method.