Tacit Knowledge and Public Accounts

Tacit Knowledge and Public Accounts

STELLA GONZALEZ ARNAL AND STEPHEN BURWOOD

The current quality assurance culture demands the explicitarticulation, by means of publication, of what have beenhitherto tacit norms and conventions underlying disciplinarygenres. The justification is that publication aids studentperformance and guarantees transparency and accountability.This requirement makes a number of questionableassumptions predicated upon what we will argue is anerroneous epistemology. It is not always possible to articulatein a publishable form a detailed description of disciplinarypractices such as assessment. As a result publication cannotachieve its stated goals. There are always elements of ourknowledge that cannot be linguistically articulated.

Dialogue between a Student (S.) and the Ideal Mathematician (I.M.).

S. Sir, what is a mathematical proof?

I. M. You don’t know that? What year are you in?

S. Third-year graduate.

I. M. Incredible! A proof is what you’ve been watching me do at theboard three times a week for three years! That’s what a proof is. [ . . . ]

S. I’ve seen arguments in geometry and algebra and calculus that werecalled proofs. What I’m asking you for isn’t examples of proof, it’s a definitionof proof. Otherwise, how can I tell what examples are correct? [ . . . ]

I. M. Well, it’s an argument that convinces someone who knows thesubject.

S. Someone who knows the subject? Then the definition of proof issubjective; it depends on particular persons. Before I can decide ifsomething is a proof, I have to decide who the experts are. What does thathave to do with proving things?

Journal of Philosophy of Education, Vol. 37, No. 3, 2003

r The Journal of the Philosophy of Education Society of Great Britain 2003. Published by Blackwell PublishingLtd., 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.

I. M. No, no. There’s nothing subjective about it! Everybody knowswhat a proof is. Just read some books, take courses from a competentmathematician, and you’ll catch on.

S. Are you sure?

I. M. Well—it is possible that you won’t, if you don’t have any aptitudefor it. That can happen, too.

S. Then you decide what a proof is, and if I don’t learn to decide in thesame way, you decide I don’t have any aptitude.

I. M. If not me, then who?(Davis and Hersh, 1980, pp. 39–40)

In The Postmodern Condition Jean-Francois Lyotard warned us to ‘expecta thorough exteriorization of knowledge with respect to the ‘‘knower,’’ atwhatever point he or she should occupy in the knowledge process’(Lyotard, 1984, p. 4). Lyotard thought that this ‘exteriorization’ wouldtake a particular form, defined by both its effects and the general methodof its attainment. Two of the key effects have been the gradualcommodification of knowledge and the end to the idea that knowledgeand learning should be valued for their own sake. The method has beenreductive: knowledge is reduced to explicitly stated packets of informa-tion. Only knowledge that is susceptible to being cashed in these termswill be valued (its value residing in its operational utility and itsmarketability) and ‘anything in the constituted body of knowledge that isnot translatable in this way will be abandoned’ (ibid.). Such notions have along history and several intellectual roots, though Lyotard foresaw wherethey were leading in our own time. Nearly twenty-five years on, we cansee the results of this trajectory. The idea that knowledge may beembodied in communities or practices and that, as such, and to asignificant degree, it is something unarticulable (a notion, in any case,marginal to the dominant epistemological paradigm in Western Europeanthought), has now been almost completely forsaken.

In recent years the demand for the explicit codification of knowledgehas served a number of inter-related (and often embedded) agendas. It isuseful to briefly rehearse these, as educational developments never occurin a social and political vacuum. In the UK these demands have beencentral to what is largely a governmentally driven programme of micro-management of the public sector. At the level of implementation it hasoften been linked to notions such as transparency, which itself has becomethe keystone in a strategy of securing accountability and quality assurancein the provision of public services. In higher education this has meant notonly that an institution’s structures and procedures have had to be madesubject to outside inspection, but that increasingly detailed aspects of theprocess of learning and teaching have had to be unpacked for scrutiny.What are largely tacit practices have required ever more meticulous and

378 S. G. Arnal and S. Burwood

r The Journal of the Philosophy of Education Society of Great Britain 2003.

explicit articulation, a task that professional practitioners themselves finddifficult and uncongenial. It is no longer sufficient for academics to beable to demonstrate or show the norms and rationales that infuse theirdecisions and methods; accountability and transparency require that theymust now be able to state what these are in a publishable form.

A common justification offered for opening up these processes in thisway is that explicitness aids ‘student centredness’ in an environmentwhere this is seen as important in advancing the student’s educationalneeds. However, while ‘advancing the student’s educational needs’ maymean that explicitness is intended to help students gain purchase oninstitutional procedure and disciplinary practice (and, in doing so, counterwhat can often be an alienating experience for many students), more oftenthe phrase simply means ‘results in improved student performance’ withina particular disciplinary genre. Furthermore, ‘student centred’ may oftenamount to little more than ‘market orientated’; and where, at one time,managers would have shied away from making this link, nowadays it isbrazenly paraded. The reason is the rise of another agenda served bytransparency: a neo-liberal ‘modernisation’ programme typified by theencroachment of market values such as ‘consumer choice’ (explicitnessfacilitating, for example, the labelling of courses—so necessary forinformed consumer choice: the ‘this module does exactly what it says onthe tin’ approach to education).

Of course, the transformation of knowledge into an explicit, publishableform also marks our transition into an economy where knowledge isconsidered the most important economic asset: the so-called ‘knowledgeeconomy’. In fact, Peter Drucker, an early theorist of the knowledgeeconomy, suggests that the reduction of inexplicit or tacit modes ofknowledge to explicit, codified forms is the defining characteristic of thisnew mode of economic organisation (Drucker, 1993). The knowledgeeconomy favours only one type of knowledge; that which is technical,codifiable, explicit and publishable—knowledge that is, above all else,patentable. It is also knowledge, so the dreamers dream, that has one otherimportant characteristic, relevant to education: it need not be learnt byapprenticeship to a master, but can be learnt in a form of training that isopen to mechanisation. Expensive, holistically fashioned, professionalpractice is thus replaceable, ex hypothesi, by cheaper, atomised, lower-order activity.

It is little wonder, then, that professional practitioners view suchdevelopments with suspicion. A reluctance on the part of faculty to spellout in explicit terms the fundamental structures of their disciplinary andpedagogical practices is invariably seen by proponents of transparency asa retreat into doctrinaire obfuscation by reactionary ‘forces of con-servatism’, intent on opposing shiny new principles of scientific manage-ment in education. Like the Ideal Mathematician, academics often respondto seemingly reasonable requests for straightforward descriptive accountsof their practices with flannel and obscurantism. And yet, despite the factthat an encounter such as that between the Student and the IdealMathematician probably is a rather frustrating experience for the Student,

Tacit Knowledge and Public Accounts 379

r The Journal of the Philosophy of Education Society of Great Britain 2003.

there is something right about the way the Ideal Mathematician resists theStudent’s overtly Socratic line of questioning—demanding a preciseformulation rather than apparently question-begging references to actualexamples. The truth is that the Ideal Mathematician has graspedsomething important about educational experience that the Student hasyet to learn: that it is not about definitions and explicit descriptions, butabout direct engagement with the constitutive practices. And this alsomeans that, pace the gurus of the knowledge economy, it must oftenconsist in the neophyte’s apprenticeship to someone with expertise in thepractice.

The issue of increasing transparency and explicitness raises interestingphilosophical questions about the nature of knowledge and learningprocesses. In particular, it makes a number of interrelated assumptions thatare worth examining. Fundamentally, it assumes that the knowledge thatgrounds a disciplinary genre1 (which includes not just appropriate stylesand subject specific facts, but also things such as relevant social norms andconventions, expectations, assumptions, academic mores, modes ofaddress, values, rhetorical requirements, the ‘folklore’ of a discipline,forms of embodied understanding, etc.—which are sometimes referred tocollectively, though inadequately, as its ‘ground rules’) is of a kind thatcan be expressed propositionally and thus stated in publishable form.Second, proponents appear to believe that doing so helps ensure thequality of provision—because such a process of ‘exteriorization’ makesjudgements publicly grounded and thus objective—and that it is the onlyway of achieving this. It also assumes, as we have already said, thathaving such knowledge available for students in a publishable form (evenif this were possible) is developmentally beneficial.

In this paper we see our task as unsettling these assumptions and, moregenerally, offering resistance to the idea that all knowledge can beexplicitly codified. At root, the insistence on publishing these ground rulesrests on a fallacy: if you cannot tell, then you do not know. We contendthat this is not so and argue that the reluctance of faculty linguistically tocodify core aspects of their practices is not always the product of anin-built professional conservatism, but may arise from the fact that thereare very real limits as to how far these tacit practices can be articulated inthe explicit form required by current quality assurance methods.Furthermore, it is not at all clear, in any case, that simply publishingde-contextualised maxims does or can play a helpful role in the inductionof students into their respective disciplinary genres. We shall focus onone particular area in which explicitness is increasingly required, thespecification of assessment criteria.

QAA REQUIREMENTS AND FUZZY CRITERIA

We should make it clear that we are not against explicitness per se. In fact,we believe that opening up to students the ground rules of disciplinarygenres can be made to serve an altogether different agenda to those

380 S. G. Arnal and S. Burwood

r The Journal of the Philosophy of Education Society of Great Britain 2003.

adumbrated above. We agree that it can play an important role in de-mystifying the educational process and, more importantly, may help re-appropriate devalued notions such as student empowerment (Burwood,1999). Rather, we will argue that the type of explicitness (its degree andmanner) that is now commonly required on teaching quality audits, andwhich is promoted in the UK by the QAA (The Quality Assurance Agencyfor Higher Education), neither helps ensure the quality and transparency ofthe assessment process, nor helps students to be more clear about what isexpected of them. The type of explicitness that we find undesirable ismade clear in the QAA’s Code of practice for the assurance of academicquality and standards in higher education. In Section 6 it is urged that‘The principles, procedures and processes of all assessment should beexplicit, valid and reliable’ (QAA, 2000, p. 5). What this means in practiceis spelt out in Section 7 and the second appendix, where it is required thatthe criteria for the marking and grading of assessments are published:‘Institutions should publish, and implement consistently, clear criteria forthe marking and grading of assessments’ (QAA, 2000, p. 8). It alsosuggests that, among the documentation on assessment that institutionsshould consider publishing, the following is specified: ‘the criteria forassessment, including, where appropriate, descriptors of expectedstandards of student attainment: what is expected in order to pass or togain a particular grade or classification’ (QAA, 2000, p. 16).

At first blush, there seems to be nothing in all this that is particularlyundesirable: it seems merely to suggest that the assessors make clear howand when a student’s work will be assessed. It is perfectly reasonable toexpect faculty to make public how many essays or exams a student willhave to do to pass a module, or when assignments will have to besubmitted. It is also reasonable for faculty to give students guidance onwhat their grades signify, and part of this process may include listing thesorts of things one looks for in, say, an essay. This is not something wewish to challenge: students need to be clear about what is expected ofthem. Explicitness, however, does not always serve clarity. Being explicitmay be seriously misleading—something that is especially true if thenature of the practice concerned is distorted by the attempted method ofensuring explicitness. The main problem with current quality-assurancepractice, however, is that it equates explicitness exclusively withpublication.

Quality audit of university teaching is largely a document-drivenprocess, so it is understandable that there is an exhortation by the QAA topublish. An obvious danger, however, is the tendency towards thecodification of criteria into sets of necessary and sufficient conditions. Buteven where this extreme is avoided, the QAA guidelines are ofteninterpreted as requiring strict descriptors; in pursuit of so-called ‘sharp’criteria instead of ‘fuzzy’ norms, which are generally considered inferior.There are several reasons why such a list of sharp descriptors is artificialand misleading and why fuzzy rules of thumb may be all that areachievable. First, all published lists tend to give the impression that theyare determining in the assessment process, when, as we will argue, they

Tacit Knowledge and Public Accounts 381

r The Journal of the Philosophy of Education Society of Great Britain 2003.

are not and cannot be so. On the one hand, they may preclude thedevelopment of a critical awareness in students who are led to believe thattheir work should mechanically conform to these maxims, which is not thecase. On the other hand, practised markers do not mark from such a list toa particular grade. Rather they make a tacitly formed judgement first andthen may refer to such a list as a post hoc justification of that judgement.There is nothing inappropriate in this approach on the part of the practisedexaminers, nor does it mean that such judgements are subjective; but itdoes require considerable latitude with respect to any list of descriptors.Lists of sharp descriptors, which are presented as determining, and whereinterpretation is supposed to be kept to a minimum, are thus even morelikely to mislead. Second, we cannot ignore the question of interpretation,where context is all. A ‘one size fits all’ list of linguistic descriptors issimply not possible; neither across subject areas, nor even within a singlesubject area. Consider just one common grade indicator for essays:‘independent thought’. Evidence of this is usually required for higher-grade bands, but it is not obviously either sufficient or necessary. It is notsufficient because we do not reward tirades of unsupported opinions,whether we think it is necessary depends on what we would count asevidence of independent thinking. Would we, for example, allow a carefuland detailed, one might even say original, exegesis of an extremelycomplex text to count?

The difficulty here is not simply one of complexity, to be resolvedby more precise linguistic formulations of the published descriptors. It isnot possible to publish ‘clear criteria for the marking and grading ofassessments’, if what this means is the codification of assessment practiceswith sharp criteria. The practice resists precise codification, so thatlinguistic formulations will always remain extremely fuzzy. Royce Sadlerput it well when he argued that ‘fuzzy standards cannot be transformedinto sharp standards by using more detailed or elaborate language, formuch the same reason that there are practical limits to the degree ofimprovement that can result from using a magnifying glass on a blurredphotograph’ (Sadler, 1987, pp. 205–206). The limitations are to do withthe nature of the practice itself and are epistemological rather than merelypractical. Our principal objection to the QAA’s requirement of explicit-ness through the publishing of detailed descriptions of criteria is that thisis based on a model of knowledge that ought to be resisted and that is, atits core, false. Assessment consists in the exercise of an applied skill, andthere are core aspects of this knowledge practice that cannot be capturedby a mere propositional description of them, thus making themunavailable for publication.

ASSESSMENT AS PRACTICAL AND TACIT KNOWLEDGE

Our main contention, then, is that there are key elements of assessmentthat cannot be expressed in the form of maxims, and that require theexercise of professional judgement. The development of the specialist

382 S. G. Arnal and S. Burwood

r The Journal of the Philosophy of Education Society of Great Britain 2003.

knowledge and its application in which such judgement consists is like thedevelopment of connoisseurship: it can only be acquired through repeatedengagement in the appropriate practices. Academic assessors do not learnwhich assessment criteria are appropriate for any given piece of work inany particular context by looking them or their specific applications up ina book and then mechanically applying them: they have to be acquired bypractice. This is a claim that rests on broader epistemological considera-tions and needs to be supported further. We will do so by introducinga pair of epistemological distinctions: between ‘technical knowledge’and ‘practical knowledge’, and between ‘tacit knowledge’ and ‘explicitknowledge’.

The first distinction was introduced by Michael Oakeshott.2 Theepistemological trend predicted by Lyotard was also identified byOakeshott, though he attributed it to what he called the ‘rationalisttradition’. Oakeshott argued that this tradition is characterised by aprivileging of ‘technical knowledge’, or knowledge of a technique, over‘practical knowledge’. Technical knowledge can be formulated into rulesand is susceptible of precise formulation in maxims. It is, therefore, thetype of knowledge that can be easily gleaned from a book, acquired byrote, learned by heart and applied mechanically. What Oakeshott termed‘practical knowledge’, on the other hand, is not susceptible to expressionin the form of maxims and exists only in use. Significantly, therefore, itcannot be learned from a book or acquired by rote, and instead can only beacquired by apprenticeship to a master. He claims that ‘These two sorts ofknowledge, then, distinguishable but inseparable, are the twin componentsof the knowledge involved in every concrete human activity’ (Oakeshott,1992, p. 12). Oakeshott resists the ‘if you cannot tell, then you do notknow’ fallacy and criticises the rationalist tradition because it assumes thatonly technical knowledge counts as knowledge ‘in the proper sense’, andbecause it marginalises the practical elements present in all knowledgeclaims. It achieves this marginalisation chiefly, he suggests, in the way itassociates rationality exclusively with codified forms of knowledge. So,for example, unless a judgement is made with reference to explicitly statedcriteria, it is, at best, subjective and, at worst, irrational (cf. GonzalezArnal, forthcoming).

It is difficult not to believe that the QAA guideline document, and thegeneral culture of accountability it represents, is motivated by these sortsof concerns: that unless regulated—this is its implication—the exercise ofprofessional judgement in education amounts to little more than arbitraryvalue judgements. Because the QAA Code of practice emphasises themaking of assessment criteria explicit through publication, without takinginto account any of the other elements present in the practice of assessing,it too seems to favour technical knowledge. Nonetheless, assessment is, aswe say, an applied skill and is an example of practical knowledge: theassessor’s knowledge exists in the practice of the skill and not in a set ofpublished maxims. There is no algorithm available for marking studentassignments or essays, and yet quality audit culture increasingly behavesas if this were not so. The privileging of technical knowledge leads to

Tacit Knowledge and Public Accounts 383

r The Journal of the Philosophy of Education Society of Great Britain 2003.

what Oakeshott describes as a ‘false but plausible theory of education’. Itis worth quoting him at length:

we are apt to believe that in order to teach an activity it is necessary tohave converted our knowledge of it into a set of propositions—thegrammar of a language, the rules of research, the principles of experimentand verification, the canons of good workmanship—and that in order tolearn an activity we must begin with such propositions. It would befoolish, of course, to deny that this device has a pedagogic value. But itmust be observed that, not only are these rules, etc., these propositionsabout the activity, an abridgement of the teacher’s concrete knowledge ofthe activity (and therefore posterior to the activity itself), but learningthem is never more than the meanest part of education in an activity(Oakeshott, 1992, pp. 111–112).

Oakeshott complained that education is increasingly regarded as the‘acquisition of a technique’ that can be obtained at a distance.3 However,what cannot be imparted in this way are what he calls the ‘nuances’ of atradition—the sort of knowledge that cannot be taught other than byworking alongside an expert practitioner (p. 39). Like Oakeshott, we wishto argue that there are core aspects of the teacher’s concrete knowledgethat cannot be converted into sets of propositions. We will even go on toargue, on pain of appearing foolish, that (insofar as it is possible)publication of such sets of propositions have an extremely limitedpedagogic value. What considerations lead us to argue, in common withOakeshott, that there are elements of knowledge that exist in a practicethat cannot, in principle, be codified? To establish this conclusion, weneed to turn to our second distinction, between tacit and explicitknowledge.

Knowing always has aspects that are tacit. This is true of allknowledge, but it is most clearly seen in practical knowledge. It issometimes assumed that what is tacit could nevertheless be made explicit;this is not always so. The distinction between tacit and explicit forms ofknowledge is Michael Polanyi’s. He postulated the existence of twodifferent types of awareness that function simultaneously in our relation-ship with the world, but which cannot be attended to at the same time; andwhich structure all experience in a way similar to the from-to structure of aGestalt. He variously terms these two types of awareness, respectively,‘explicit’ and ‘tacit’, ‘distal’ and ‘proximal’, ‘focal’ and ‘subsidiary’(Polanyi, 1969, pp. 138–158) and offers the following example toillustrate them: when we use a hammer to drive in a nail, we attend to bothnail and hammer, but in a different ways (Polanyi, 1998, p. 55). Our ‘focalattention’ is on the nail, as it is going to be hit; but to be able to do this, heargues, we have to be ‘subsidiarily aware’ of the hammer as it is beingheld by us. We attend to the nail from the hammer and from other tacitclues and structures. Our focal attention allows us to direct our efforts tothe realisation of the task in hand, by giving us a general feeling for thesituation. The elements of which we are subsidiarily aware act as clues

384 S. G. Arnal and S. Burwood

r The Journal of the Philosophy of Education Society of Great Britain 2003.

that allow us to make sense of what we are focally aware of and, at thesame time, these clues only become meaningful in relation to the whole.The action of going from what we are subsidiarily aware of to what is inour focal attention is an act of integration, and this act always remainstacit.

Polanyi claimed that explicitly stated ‘rules of the art’ cannot replacepractical knowledge, and explained why there is some knowledge thatalways remains tacit. First, integration itself is a skill for which a completedescription in the form of maxims cannot be given. We become skilled inascertaining which parts of the world are salient for our performance andwhich are not. This is a skill that is learned socially, and that takes theform of knowing how to perform an activity. We remain unaware that wehave taken decisions when ascertaining how to hit the nail, and cannot tellhow we have taken these. Although we cannot sum up the steps that wehave taken to reach the end result, the decisions we take are shown in ourperformance. Becoming able to grasp new patterns, to understand newpractices, to make new integrations, is a social process of acquiring skills.Once we have acquired them, we put them into practice without thinkingabout it and they become second nature. Polanyi describes this process ofacquiring a second nature as ‘dwelling in the knowledge’, and compares itwith the way in which, by using tools, we perceive the world through themas if they were an extension of our body. By using them, we indwell inthem; we accept them existentially (Polanyi, 1998, p. 59). This does notmean that our grasp of any underlying principles is subjective, or that theirapplication is thereby arbitrary, as normativity is preserved in thecommunal practice. While it is true that the application of the principlesowes something to intuition, judgements are open to justification andchallenge (and, tellingly, the intuition of a beginner does not command thesame respect as the intuition of a master).

Polanyi also makes a clear distinction between possible descriptions ofthe principles that underlie a practice and what is tacitly known by anagent. What we are subsidiarily aware of can sometimes be brought intoour focal attention, so we can formulate it in the form of maxims; but whatis tacitly known are mere clues in relation to a whole, and are meaningfulonly in relation to it. There is an interpretative element in makingintegrations that cannot be captured by an atomic analysis of the differentelements that are subsidiarily known. In making decompositional analysesof the elements present in the from-to relationship, we cannot capture therelationship itself and the dynamic elements present in our integrations.For instance, we can explain the general principles followed by riderswhen they ride a bicycle, but knowledge of these, as a theory, does notimply that one can ride a bicycle, and in actual fact, they do not even aidperformance (Polanyi, 1998, p. 50). A theory of how to ride a bicycle isnot usable by a cyclist to help her improve her performance because theelements that act as clues in her subsidiary awareness when she is riding abicycle and the theory are diverse. So, even if we can publish certainknowledge in a manual, codification will always remain fuzzy, as it isimpossible to specify how this explicit knowledge relates to concrete

Tacit Knowledge and Public Accounts 385

r The Journal of the Philosophy of Education Society of Great Britain 2003.

instances of the practice. These are particulars of a different kind and theirrelation can be existentially apprehended only in use.

How is all this relevant to our discussion of assessment criteria? Afterall, examples such as hammering and cycling are examples of embodiedagency. For Polanyi, our embodiment is an important tacit factor as itinfluences all our relationships with the world, and mediates all ourknowledge. We should not be too hasty in assuming the subsidiaryawareness of our embodiment is irrelevant, even in abstract intellectualtasks.4 We are generally not aware of the many ways in which ourembodiment mediates our knowledge. Furthermore, in the same way inwhich we assimilate actual tools existentially, we also assimilateintellectual tools, such as languages, scientific theories or even moralteachings. We do this by participating in the relevant social practices,observing, copying, imitating, until we begin to grasp the sense of theactivities and are able to integrate different elements. Academic assessmentis a form of practical knowledge that has tacit, unspecifiable core aspects.The host of factors that go to make up the ground rules of a disciplinarygenre act as tacit clues in our subsidiary awareness. These clues are onlymeaningful within a context, relationally, and became unusable if we applythe method of analysis to them. They are existentially apprehended in theirapplication, and are thus different in kind from ‘objective’ descriptionsoffered in the form of rules. Their application requires an act of integrationby a practised assessor who indwells in the knowledge, and whosejudgements are normatively constrained in the practice. The alternativemodel is to postulate an impossible infinite regress of codified knowledge(see Ryle, 1946; 1949). For example, any list of sharp criteria would needto be accompanied by another list on how to apply them (and would bemeaningless without it) and this, turn would require another list, and so on.Finally, the way in which we are socially induced into disciplinary genressupports the idea that we make sense ‘existentially’ of the constitutivepractices first, before we can reflect on them.

ACCULTURATION INTO A PRACTICE

Because the application of a published list of underlying principles cannotitself be codified, the pedagogic utility of such lists is extremelycircumscribed. Addressing explicitness by means of presentationalapproaches assumes that the mere publication of assessment criteria byitself is beneficial in the development of the students’ understanding; thatis, it implies that when students read published criteria they will knowwhat is expected from them. It is just not obvious that, when confrontedwith a set of necessarily fuzzy and de-contextualised norms in this way,prior to engagement with the relevant practices, the neophyte will be ableto grasp their significance. What is worse, instead of helping students tobecome inducted into the practices, which would allow them to apprehendwhat makes a good piece of assessed work and how to produce one, theyare provided with a list of propositions that ineffectively substitute othertypes of schema for induction. Induction may be more time-consuming,

386 S. G. Arnal and S. Burwood

r The Journal of the Philosophy of Education Society of Great Britain 2003.

but it would provide the students with a genuine purchase on the practice.It is not that the ground rules of a disciplinary genre cannot be madeexplicit; they can, but from within the practice.

The instrumental effectiveness of teaching explicitly is still acontentious issue in empirical educational research, though the sorts ofepistemological reflections we have explored should give pause forthought to its advocates. Where explicit instruction is common (forexample, in sports coaching or in training on a musical instrument)explicitly acquired skills must become tacit as part of a from-to structurein order to be used productively. More importantly, the consciousappraisal of one’s performance is always disruptive of these structures.Thus, in The Inner Game of Tennis, William Gallwey notes the fact thatthe lament ‘Oh, I can never do anything I try to!’ expresses an importanttruth; viz. deliberately trying to do something disrupts the smooth flow ofa spontaneous performance (Gallwey, 1974, p. 27). Precisely the samepoints apply to academic study. For example, it is not uncommon toencounter a student who is unable to write an essay simply because she isso concerned with what she has to do to write an essay. She has become, inone of Polanyi’s own examples, like the speaker who becomes inarticulatethrough listening to his own voice and concentrating on the movement ofhis own tongue and lips. In order for her to use creatively any explicitinstruction she has received, she needs, in effect, to forget it and immerseherself in the act.5

An emphasis on publication also ignores the transformative nature oflearning. The process of indwelling changes our nature, mediating ourrelationships to the world and others (Scott, 1971). Thus, it is not anexaggeration to say that education is about acquiring a new identity (seeHarre, 1983; Wenger, 1998). In higher education the process through whichthis is achieved is by means of a gradual, and largely tacit, acculturation orinitiation of a student into a discipline. The information supposedlycontained in published maxims is not available to those as yet uninitiatedinto the relevant practices. Acculturation is an active process that necessarilybegins with a submission to authority; mainly because the neophyte is notable to judge the appropriateness or inappropriateness of behaviours withinpractices until she has became able to see its coherence. It is a process thattherefore requires a degree of trust from the student and consists inparticipation in anticipation of understanding. As Polanyi says, by:

watching the master and emulating his efforts in the presence of hisexample, the apprentice unconsciously picks up the rules of the art,including those which are not explicitly known to the master himself.These hidden rules can be assimilated only by a person who surrendershimself to that extent uncritically to the imitation of another (Polanyi,1998, p. 53).

The way in which we are inducted into practices thus adds a further reasonwhy there are unspecifiable contents in all knowledge. We learn primarilyby imitating, by copying the ways of those who have mastered (or are

Tacit Knowledge and Public Accounts 387

r The Journal of the Philosophy of Education Society of Great Britain 2003.

mastering) the practice already, and we are able tacitly to pick up rules inthe behaviour of those with expertise that remain unknown to them.Polanyi would probably endorse the seemingly unhelpful responses of theIdeal Mathematician in Davis and Hersh’s dialogue, as he insists that wecan only learn mathematical theory through practice. It is through practice,for example, that we learn to recognise that a particular puzzle is just aninstance of a more general type. Mathematicians have to undergo a certaintraining that allows them to develop skills that make them able to seethings that would not be meaningful or salient to a less-trained eye. Theyare able to make an integration of knowledge that remains tacit and aconclusion that becomes explicit. It is a fact that, if we ask people howthey know that two puzzles are instances of a more general type, they,generally, would not be able to tell. Once we reach a certain level ofcompetence, and indwell in the practices, they become second nature, andwe then see the world as mediated by them. But, as well as acquiring thepractices, we also became able to judge our performances and those ofothers. This critical stance is only inculcated by participation in thepractice, given that it is the only way of grasping its coherence, and notwith reference to externalised maxims. It is therefore necessary toparticipate in order to grasp the normativity of the practices. This mightnot be specifiable, but it is, nevertheless, graspable.

All this suggests that, if we want students to gain a real purchase on theunderlying principles of our disciplinary genres, we need to encouragedevelopment of connoisseurship in them as well. For this, there is reallyno substitute for repeated engagement in the appropriate activities in thecompany of those with expertise. However, such practice does not have tobe blind, as more appropriate, experientially based induction schemas arereadily available. Among these are self- and peer-assessment, commentson essays, the use of exemplars, drafting, talking to peers and tutors aboutassessment, suggesting their own essay titles and offering rationales, aswell as practising different writing styles (for example, in philosophy anexcellent alternative to the normal essay is the dialogue). This type ofengagement with assessment practices would not make the criteria explicitin the ways suggested by the QAA, given that we think that is impossible,but would help the student to apprehend them, to be clear about them,develop a critical way of thinking about them, and to be able to engage inthe practice more appropriately and hence more effectively (see Race,2001). As part of such activities (and only as part of them) published listsof fuzzy norms may play a role, but they are not necessary. Students canbe made to reflect critically on their own performance, without referenceto published lists of explicitly stated assessment criteria.

CONCLUDING REMARKS

Education is, as we say, a process of participation in anticipation. Strivingfor ever-more detailed published descriptions not only does not help inthis process: it could actually damage it, as it may undermine theappropriate relationship between faculty and students. As Onora O’Neill

388 S. G. Arnal and S. Burwood

r The Journal of the Philosophy of Education Society of Great Britain 2003.

has recently claimed, the production of data does not increase eitheraccountability or trust, and, in actual fact, may have the opposite effect,producing uncertainty and mistrust (O’Neill, 2002). In fact, one of theironies of the current quality assurance culture is that it cannot, perimpossibile, even achieve what it is meant to achieve, accountability andthe assurance of quality. This is because, as we have shown, providingexplicit publishable data by itself does not amount to giving relevant,usable information to students and other interested parties, as publicationdoes not disclose what is involved in the practice—in this case, that ofevaluating and assessing a piece of work. Academics should be trusted intheir ways of assessing students, not because they adhere to an‘exteriorized’, and distorting, published description of how they do this,but rather because they can engage with the students in such a way as tomake them understand how it is done, because they are able to engagewith their peers, and because they can show, if not state, the norms andrationales that infuse their decisions and methods.6

The Ideal Mathematician’s response to the Student was thereforeabsolutely correct, it is not subjective, but it does depend on particularpersons—a response that occasionally leads advocates of codification toimply that this model of knowledge and learning is anti-democratic (seeDrucker, 1993, Chapter 1). Yet explicitness does not guarantee inclusionor democracy, especially if all this means in the educational context is thatfaculty reveals otherwise tacit but non-negotiable criteria (see Burwood,1999). The transformative nature of education does not mean it is anti-democratic, but it does means it is exclusionary (especially given itsrelation with personal identity). The tacit knowledge of underlyingprinciples is inaccessible to those outside the practice (and, in our case, itis thus only available to academics and students with the relevantexperience in their disciplines). But why should it be thought otherwise?Knowledge and learning just are exclusionary in this sense and cannot behad on the cheap or acquired via short cuts. Unless we are to allow whatOakeshott termed ‘the meanest part of education in an activity’ to dictatethe conduct of higher education as a whole, we would do well toremember Aristotle’s dictum: ‘We must be content, then, when speakingof such subjects . . . to indicate the truth roughly and in outline . . . for itis the mark of the educated man to look for precision in each class of thingjust so far as the nature of the subject admits’ (Aristotle, 1925, p. 3).7

Correspondence: Stella Gonzalez Arnal and Stephen Burwood, Depart-ment of Philosophy, University of Hull, Cottingham Road, Hull, HU67RXEmail: [email protected] and [email protected]

NOTES

1. We understand ‘genre’ to indicate a particular form of social action, rather than simply a particular

style (cf. Miller, 1984).

Tacit Knowledge and Public Accounts 389

r The Journal of the Philosophy of Education Society of Great Britain 2003.

2. In fact, this distinction has been made several times under different guises: ‘know-how’ and

‘know-that’ (Dewey, 1922); ‘knowing-how’ and ‘knowing-that’ (Ryle, 1946); ‘declarative

knowledge’ and ‘procedural knowledge’ (Anderson, 1983); ‘propositional knowledge’ and ‘non-

propositional knowledge’ (Johnson, 1987). The distinction between ‘tacit knowledge’ and

‘explicit knowledge’ (Polanyi, 1969), on which we also draw, is related though significantly

different. See also ‘participation’ and ‘reification’ (Wenger, 1998)—though Wenger would

maintain that this is fundamentally a different type of distinction as its components are not

classificatory or mutually exclusive. In point of fact, both Oakeshott and Polanyi make it quite

clear that they think of their distinctions as dynamic relations composed of inseparable

constitutive elements.

3. He was specifically concerned with professional education, though the point applies more

generally, and certainly to higher education. If what he says was true at the time he wrote it (the

late 1940s), how much more true it is today? An interesting question this point raises, which we do

not have space to address, is ‘What counts as ‘‘at a distance’’ or ‘‘working alongside an expert

practitioner’’?’ Oakeshott clearly thought that embodied presence is important and our inclination

is to agree, especially if, as Hubert Dreyfus says, we want to impart expertise and practical

wisdom rather than just competence (Dreyfus, 2001, Chapter 2).

4. We have a severely dyslexic student who recently produced an excellent essay, which was clearly

written and well structured, to a degree well above his usual level. When he was asked what he

thought had made a difference, he suggested that it might have been that his room had been clean

and tidy. Who of us does not feel inspired in the right surrounds or inhibited in the wrong ones?

5. The best techniques to help a student in this predicament, such as free-writing, in fact have little or

nothing to do with criteria that reflect disciplinary practice.

6. Polanyi is right to emphasise that acculturation primarily requires the formative presence of the

expert practitioner as a role model, though we should also recognise the important fact that

students also learn tacitly from each other (see Gerholm, 1990).

7. Stella Gonzalez Arnal would like to acknowledge the Fundacion Seneca, who funded the period of

her research during which this paper was written. Both authors would like to thank colleagues for

their helpful comments on earlier drafts, especially fellow participants at the Society for Applied

Philosophy 20th Annual Conference, ‘Education and Society’, Mansfield College, Oxford, 28–30

June 2002.

REFERENCES

Anderson, J. R. (1983) The Architecture of Cognition (Cambridge, MA., Harvard University

Press).

Aristotle (1925) The Nichomachean Ethics, trans. D. Ross (Oxford, Oxford University Press).

Burwood, S. (1999) Liberation Philosophy, Teaching in Higher Education, 4.4, pp. 447–460.

Davis, P. J. and Hersh, R. (1980) The Mathematical Experience (Harmondsworth, Penguin).

Dewey, J. (1922) Human Nature and Conduct. An Introduction to Social Psychology (London,

Allen & Unwin).

Dreyfus, H. L. (2001) On The Internet (London, Routledge).

Drucker, P. F. (1993) Post-capitalist Society (Oxford, Butterworth Heinemann).

Gallwey, W. T. (1974) The Inner Game of Tennis (London, Jonathan Cape).

Gerholm, T. (1990) On Tacit Knowledge in Academia, European Journal of Education, 25.3, pp.

263–271.

Gonzalez Arnal, S. (forthcoming) Rationality, Normativity, and Practices. Towards a More

Inclusive Concept of Reason, Women: A Cultural Review, 14.

Harre, R. (1983) Personal Being (Oxford, Basil Blackwell).

Johnson, M. (1987) The Body in the Mind: The Bodily Basis of Meaning, Imagination, and Reason

(Chicago, Chicago University Press).

Lyotard, J. -F. (1984) The Postmodern Condition: A Report on Knowledge (Manchester,

Manchester University Press).

Miller, C. (1984) Genre as Social Action, Quarterly Journal of Speech, 70, pp. 151–167.

O’Neill, O. (2002) The Reith Lectures 2002: A Question of Trust: online at http://www.bbc.co.uk/

radio4/reith2002/thelectures.shtml

390 S. G. Arnal and S. Burwood

r The Journal of the Philosophy of Education Society of Great Britain 2003.

Oakheshott, M. (1992) Rationalism in Politics and Other Essays (Indianapolis, Liberty Press).

Polanyi, M. (1969) Knowing and Being, ed. M. Grene (Chicago, University of Chicago Press).

Polanyi, M. (1998) Personal Knowledge: Towards a Post-Critical Philosophy (London,

Routledge).

The Quality Assurance Agency for Higher Education (2000) Code of practice for the assurance of

academic quality and standards in higher education, Section 6: Assessment of students

(Gloucester, QAA).

Race, P. (2001) LTSN Assessment Series No. 9: A Briefing on Self, Peer and Group Assessment

(York, LTSN Generic Centre).

Ryle, G. (1946) Knowing How and Knowing That, Proceedings of the Aristotelian Society, 46, pp.

1–16.

Ryle, G. (1949) The Concept of Mind (London, Penguin).

Sadler, D. R. (1987) Specifying and Promulgating Achievement Standards, Oxford Review of

Education, 132, pp. 191–209.

Scott, W. T. (1971) Tacit Knowing and The Concept of Mind, Philosophical Quarterly, 21.2, pp.

22–35.

Wenger, E. (1998) Communities of Practice. Learning, Meaning, and Identity (Cambridge,

Cambridge University Press).

Tacit Knowledge and Public Accounts 391

r The Journal of the Philosophy of Education Society of Great Britain 2003.