Truth, Balance, and Freedom

Editor's Note: Akeel Bilgrami has kindly given us permission to publish here the text of a lecture that he delivered at The New School and which is also included in the Chomsky Notebook collection of writings.

by Akeel Bilgrami

9780231144759 Though there is much radical –and often unpleasant– disagreement on the fundamental questions around academic freedom, these disagreements tend to be between people who seldom find themselves speaking to each other on an occasion such as this or even, in general, speaking to the same audience. On this subject, as in so much else in the political arena these days, one finds oneself speaking only to those with whom one is measurably agreed, at least on the fundamental issues. As proponents of academic freedom, we all recognize who the opponents of academic freedom are but we seldom find ourselves conversing with them in academic conferences. We only tend to speak to them or at them in heated political debates when a controversy arises, as for instance at Columbia University over the promotion of faculty in Middle Eastern studies, or in those states where the very idea of a curricular commitment to modern evolutionary biology is viewed with hostility. I will not be considering such controversial cases of overt political influence on the academy. This is not because they are not important. The threats they pose are very real, when they occur, and the need for resistance to these threats is as urgent as anything in the academy. But they raise no interesting intellectual issues at a fundamental level over which anyone here is likely to be disagreed. If there is disagreement in a forum of the kind at which we are presently gathered, it is likely to be on relatively marginal questions, such as, for instance, whether academic freedom is a special case of the more basic constitutional right to free speech or whether instead it is a special form of freedom tied to the specific mission of universities.

What might a philosopher contribute to these more marginal questions? In this brief lecture, I would like to make a fuss about a standard argument for a conception of academic freedom which we all seem to subscribe to when it is coarsely described but which, when we describe it more finely, and look at the arguments more closely, is quite implausible and leads directly to thoroughly confused ideas about displaying ‘balance’ in our classrooms and our pedagogy quite generally. I will then use some of the points and distinctions I make in this critique to explore whether there is scope for locating more subtle and interesting (and actually more pervasive) kinds of threat to academic freedom than the obviously controversial ones that I mentioned above which all of us here, I assume, find an abomination, and which, as I said, raise no interesting issues for any of us, even if they ring urgent alarms. At the very end, I will venture to advocate imbalance of a very specific kind in the ‘extra-mural’ domain, when it is neither inquiry nor classroom curriculum that is at stake but the effort to engage the intellectual and political culture at large.

II

No matter which stand is taken on the marginal question as to whether academic freedom is a special case of the constitutional right to free speech or something special and apart, there is a great and recurring tendency in the literature on the subject to appeal to the same broad arguments and metaphors and intuitions to present the justifications for academic freedom as is done for the justification for freedom of speech in general. And it takes roughly the following lines. First, there is a statement of purpose or goal: academic institutions are sites for intellectual inquiry and research and therefore one of their chief goals is the pursuit of truth and the pedagogical project of conveying the truth, as one discovers it and conceives it in one’s research, to students, and to set students on the path of discovering further truths in the future on their own. And then second, there is a statement of the conditions for the pursuit of that goal: this pursuit of truth is best carried out, it is said, under conditions where a variety of opinions are allowed to be expressed on any subject, even if one finds some of them quite false, since it is possible that they might be true and one’s own view might turn out to be false. Often, the metaphor used to capture this ethos and it’s efficacies in the matter of truth, is that truth surfaces in a ‘market place of ideas’.

When Justice Holmes first put that phrase into the air, he was not particularly thinking about the academy, but quite generally about the shape of a free society.[ii] In fact, as two Columbia historians (Richard Hofstader and Walter Metzger[iii]) pointed out, Holmes was really expressing in more intuitive and metaphorical terms the justification for tolerance in speech quite generally, for which John Stuart Mill had earlier in On Liberty given a more structured argument with premises and a conclusion.[iv] So even if one thought that academic freedom was set apart from the articulations of the First Amendment, the structure of the underlying philosophical argument is the same as to be found in Mill’s more general argument for liberty of speech as a fundamental principle of the polity at large. I want to spend some time on this underlying argument but before I do, it is worth emphasizing that it is not just given by professional and lay philosophers, it is found in the case law of this country in which universities have figured, repeatedly. Thus for instance in ‘Keyishan vs Board of Regents of the State University of New York’ (1967), the language of the Supreme court of this country explicitly cites the phrase ‘marketplace of ideas’ and talks of the ‘robust exchange of ideas which discovers truth out of a multitude of tongues’. That is just one example. There are literally scores of cases in the lower courts as well that appeal to Millian considerations, and they too begin by defining the goal of universities as being one of seeking the truth in intellectual inquiry

What is Mill’s argument and why does it have such a strong appeal for law, philosophy, and even our everyday understanding of the justifications for academic freedom? Its appeal is the appeal of a certain fallibilist epistemology that widely underlies the classical and orthodox liberal mentality. Curiously, this form of fallibilism clashed starkly with the pragmatist epistemology of American thinkers like Peirce and also with the heterodox from of liberalism that one finds in American thinkers like Dewey. Yet the American courts and American quotidian opinion cite Holmes and Mill like a mantra.

Mill’s argument has two premises and a conclusion. The premises are:

Premise 1: Many of our past opinions, which we had held with great conviction, have turned out to be false.

Premise 2: So, some of our current opinions that we hold with great conviction may also turn out to be false.

From these premises, he drew a conclusion about tolerance and free speech,

Conclusion: Therefore, let us tolerate dissenting opinions just in case our current opinions are wrong and these dissenting opinions are right.

The idea is that ‘market place of ideas’ keeps us honest. Since we can never be sure that we are right, a market place of opinions, many of which may oppose our own opinions, may well throw up the truth, displacing our own convictions about it. Metzger and Hofstader make this connection between Holmes and Mill explicit and there is no doubt that something like this justification, if true, would hold for free speech in the academy with particular force, even if we saw the academy as standing apart from constitutional contexts for free speech, because the academy is specially geared to pursue the truth in its various disciplinary pursuits.

Let’s then stare at the argument for a while.

Mill’s argument is based on an induction. It is often called Mill’s ‘meta-inductive argument’. The induction is found in the transition from the first premise to the second. It is called a meta-induction presumably because whereas most inductions go from observations about the world in the past to conclusions about the future, his induction goes from an observation about our past beliefs about the world to a conclusion about our present and future beliefs (viz., that they may be false).

There is an extraordinary ambition in this argument. It hopes to persuade us of a value, the value of free speech, as something for a polity or a university to embrace, on the basis of something that is pure rational argument. By this I mean that it does not aim to convince us to adopt a value (the value of free speech) on the basis of any other moral or political values. It hopes to convince us on grounds that are, in that sense, value-free. It does not matter what moral or political values we have, so long as we are capable of induction, we are supposed to see the force of the argument. And since inductive capacities, like deductive capacities, are part of general rational capacities, possessed by all human beings, if the argument is right, everyone should see the value in free speech, just in virtue of their rationality. To fail to do so, therefore, is nothing less than irrational. Mill gives quite other arguments for free speech in that careless masterpiece –such as for instance that free speech is a value to live by because it encourages diversity as well as creativity in society, and that a willingness to submit to the clash of ideas is essential to the moral courage of human beings and prevents their mental pacification. But such argument is inherently less ambitious. Its appeal is confined to those who value individual creativity, or variety, or what Blake called ‘mental fight’. There is a risk in any argument that comes to an evaluative conclusion by appealing to another value. Values are things that have variable appeal. And so those who do not subscribe to the other value will not be convinced by it. The meta-inductive argument, by contrast, if successful, is supposed to knock us down with a much more general logical force.

But is it successful? The incessant sloganeering about the ‘market place of ideas’ depends centrally on its success. Deep though it goes in liberal culture and sensibility, I think Mill’s argument is a numbing fallacy.

To begin with, even at a cursory glance, you will notice that the judgement in the first premise is made from the point of view of one’s current opinions and convictions. It is from our present point of view, from what we currently take to be true, that we are able to say that our past opinions are false. But the judgement in the second premise is telling us that our current point of view may contain false views and therefore to be unsure and diffident about them. Now, if we are unsure about our current beliefs, and our judgement in the first premise is made on the basis of our current beliefs, then to that extent we must be unsure of our first and basic premise. Any conclusion based on it therefore is bound to be, to that extent, itself shaky and uncertain.

There is another more fundamental internal problem with the argument.

In characterizing it, I have said that it comes to a value conclusion on the basis of premises that appeal merely to an induction, and not on the basis of any other political or moral value. But the fact is that though it appeals to no moral and political values, it does appeal to a cognitive value, the value of truth. Since it says that one should adopt free speech because it creates a market place of ideas from which the truth, even if it goes against one’s convictions, will emerge, one is assuming at least that there is value in pursuing the truth. So, it does appeal to another value to justify the value of free speech. It is only because we value truth and have it as a goal that we will be moved by the idea that a market place of ideas engendered by freedom of speech is something that we should adopt.

But, now, if that is so, there is something internally peculiar about an argument that appeals to the value of truth and the goal of pursuing the truth, as it does, while also implying, as the second premise does, that we can never know that we have achieved the truth. How can we claim to have a goal that we can never know we have achieved, when we have achieved it? What sort of goal is that? It is not perhaps as peculiar as having a goal that we know that we can never achieve. That is outright incoherent. You cannot coherently strive to achieve what you know to be impossible. But to allow that we can achieve a goal and yet insist that we can never know we have achieved it when we have, though not perhaps outright incoherent, is a very peculiar understanding of what goals are.

To put it explicitly, the internal tension is this: The argument’s second premise says that beliefs whose truth we are utterly convinced about may turn out to be false. This strictly implies that we can never be sure that we have achieved the goal of truth, not even when we are quite convinced we have. And yet the argument presupposes that the pursuit of truth is a value and that we have it as a goal to pursue. If the goal of inquiry into the truth that all academic institutions embrace is really to pursue in this way something that we never can be sure we have achieved, then we must be assuming that what we do, in pursuing it, is a bit like sending a message in a bottle out to sea. We never know what comes of it, we never know that it has arrived. What sort of epistemological project is that? It is a conception of inquiry in which we have no control over its success. If inquiry is successful, that success is, from our hapless point of view as inquirers, necessarily some sort of bonus or fluke.

The argument demands that our point of view of inquiry have a built in diffidence: we are supposed to be diffident even about our most well established claims. But such diffidence yields no instruction. The doubt expressed by the thought ‘for all one knows even our strongest convictions as to what is true might be false’ is an idle form of doubt. Consider the paradox of the preface, in which the author says coyly, “Something or other that I say in the next four hundred pages is bound to be erroneous or false” (and then typically adds, and “for those errors I alone am to blame and not all those nice people I have just acknowledged as having aided my thought and argument”). The author’s declaration of impending falsity in the pages to come is idle because it gives him or her no instruction about what to do to remedy things. It is not as if he knows what it is that is bound to be false, and why. Like Mill and Holmes he just thinks that that is the tentativeness and diffidence with which we must hold the views we have written down. But a doubt that gives no instruction in his practice of writing is a doubt that does not make any epistemic difference. And as pragmatists say, something that makes no difference to practice (not even to cognitive practice, as in this case) makes no difference to inquiry and epistemology at all. [v] Any argument which arrives at a commitment to free speech on the basis of a conception of inquiry that has such precarious coherence hardly deserves the centrality that it has been given in the liberal tradition of political thought.

In the immediate context of the political controversies we find ourselves in, in university life, the conception of academic freedom based on such a classical liberal form of argument leads directly to the advice we often get, sometimes even from University presidents, about how we should be balanced in what we say in our classrooms, showing consideration to all points of view even those which from our point of view we confidently know to be wrong. This directive wholly fails to understand what sort of role the ideal of ‘balance’ ought to play in the academy. It is a worthy ideal but we have to understand the right place and context for it in the academy.

Let’s go along, as we have been doing, with the assumption that a primary aim of universities is to pursue the truth in our various disciplinary inquiries and that the point of pedagogy is to try and present the truth we have found by presenting evidence and argument for it. Now if ‘balance’ has any role to play in all this, its role is entirely nested within this primary goal, not something independent of this goal. And within this primary goal, the only thing that ‘balance’ could mean is that one must look at all the evidence that is available to one in our inquiries. (This is the cognitive counterpart to what decision theorists call ‘the total evidence requirement’.) What ‘balance’ cannot possibly mean is the nonsensical thing that the directive we are considering tells us, viz., the equal presentation in the classroom of two contradictory views. No educator with any minimal rationality would do that on the elementary grounds that if there are two contradictory views, only one can be right. Of course if she cannot make up her mind on the evidence as to which one is right, she might present the case for both views even-handedly. But presumably such undecidedness is an occasional phenomenon. If so, balance cannot be put down as a requirement for pedagogy in the classroom. Hence, the constant demand that we always present both sides of a disagreement presupposes a conception of education as a sort of chronic dithering. It is far more sensible to say that ‘balance’ allows that an educator presents her judgement with complete conviction because ‘balance’ in the academy is nothing other than a synonym for the idea that we must look at all the evidence before coming to our convictions. It has no other role or meaning . Attempts to give it another meaning (as in the directive with which I am finding fault) are drawn from a fault-line that has its beginnings in the canonical Millian form of liberal argument for free speech.[vi]

III

I have been inveighing against a very standard liberal argument and a metaphor that it yields about truth emerging from a market place of ideas, which goes deep in the sensibility of our self-understanding in the academy and in the courts that have pronounced judgement in controversial cases that the academy has thrown up. This may have given the impression that I am recommending more dogmatism regarding our own convictions than a commitment to academic freedom can allow. That impression would be wrong.

The criticisms I have just made of Mill’s argument are quite compatible with the view (which is my own view) that there is far too much dogmatism in the academy, especially in the social sciences and even in the humanities. (And if it is less so in the natural sciences, still, as Kuhn pointed out almost five decades ago, there is some there too.) As a matter of fact, my view is that if we could characterize more or less exactly what this dogmatism is, we would have identified the most pervasive as well as the most insidious and interesting form of threat to academic freedom.

As I said at the outset, this paper was going to raise a typical philosopher’s fuss about how to rigorously characterize the arguments by which we justify academic freedom and I have said that I find Holmes’s metaphor and Mill’s argument less than exact and plausible and this implies that theirs is not the way to understand the dogmatism that thwarts academic freedom. To be fussy is to demand that one gets certain distinctions carefully right. And I am claiming that to diagnose and combat the far too high levels of dogmatism in the academy, we do not have to assume a fallibilist notion of diffidence and doubt. It is one thing to be undogmatic in the way that academic freedom demands, quite another to have the sort of notion of inquiry suggested by Millian and classical liberal arguments for academic freedom.

Let me convey what I have in mind by the dogmatism that constitutes a threat to academic freedom by returning to the paradox of the preface. The paradox offers us a site for locating a useful taxonomy via which we can identify what sort of dogmatism amounts to such a threat.

I had said about the paradox that the generalized, that is to say, the unspecific form of doubt that is stated in the preface (‘something or other in what follows in these pages may not be true’, echoing Mill’s argument that our strongest convictions may turn out to be false) gives the author no instruction as to what to do about it. He cannot possible be moved to do anything about his text by a doubt such as this. What the author will be moved and instructed by is not this sort of doubt but rather –if he or she is not dishonest and not obtuse– by some specific evidence or argument that is provided against one or other of his specific conclusions or claims. Now, both these qualifications ‘if he or she is not dishonest and not obtuse’ are revealing.

They show that there is no direct relevance of this issue I have just raised (about ignoring specific counter-evidence and counter-argument presented to one) to the question of academic freedom. Suppose someone failed to recognize counter-evidence that was presented to him. That would be a sign of his obtuseness. Suppose again that someone did recognize that counter-evidence had been presented to him by some colleague and he simply ignored it. That would be a sign of his intellectual dishonesty. But both these things are quite separate kinds of wrong from thwarting academic freedom. Now, it is true that sometimes those who are dishonest in this way are caused by this dishonesty to suppress or hound out someone who presented that evidence and that would, of course, be threatening to academic freedom; but suppressing or hounding someone out is a matter quite separable from what we are concerned with, the ignoring of evidence that is provided against what one takes to be the truth.

If this is right, we have identified so far three different phenomena. First, there is academic dishonesty –to recognize evidence or argument that goes against one’s conclusions but ignore it. This in itself is not academic unfreedom. Second, there is the inability to even recognize the force of counter -evidence and counter-argument. Let’s call this academic or intellectual obtuseness. And, even more obviously, that is not a case of academic unfreedom either. Third, there is the suppression of those who present counter-evidence and counter-argument that one has recognized to be so and one has dishonestly evaded. This, I have said, is a case of academic unfreedom. But, as I said at the beginning of the lecture, it is a very obvious case and not a very interesting one, so I will simply put such cases aside since they raise no difficult questions. It is not even clearly characterizable as a case of dogmatism though it bears some relations to dogmatism.

We, then, still do not have the kind of academic unfreedom that is genuinely and clearly also a case of dogmatism. So now, finally fourth in our taxonomy, I want to present that kind of dogmatism and show why it is a far more interesting and unobvious and also a more pervasive threat to academic freedom than is identified in the third; and in presenting it, it will become clear what its relation is to the first and second phenomena in the taxonomy, from which it is also important to distinguish it, especially the first phenomenon with which it is too often conflated.

The dogmatism that interests me is found in submerged forms of academic exclusion when we circle the wagons around our own frameworks for discussion so that alternative frameworks for pursuing the truth simply will not even become visible on the horizon of our research agenda. This form of dogmatism is distinguishable from the first of our four phenomena, academic dishonesty of the kind that refuses to accept counter-evidence and argument presented in refutation of some specific conclusion of our inquiry. Why? Because alternative frameworks do not refute our conclusions directly with counter evidence or arguments, so much as point to other, possibly deeper and more interesting ways of looking at what we are studying. And here is the crucial point. If they do contain counter-arguments and counter-evidence to our own claims and convictions, those will only surface further downstream, well after the frameworks are recognized by us upstream as possibly fruitful forms of investigation. But it is this recognition upstream that the dogmatist in us finds so hard to confer and it is in this failure that academic unfreedom (rather than intellectual dishonesty) is located.

These are cases in which a discipline discourages the development of frameworks outside of a set of assumptions on which there is mainstream consensus –and the political influence on the formation and maintenance of these exclusive assumptions, where it exists, is very indirect indeed, so indirect that it would need a fair amount of diagnostic work to reveal it since the practitioners themselves are often quite innocent of the influence. (On the other hand it is not as if this is a rare or unusual phenomenon. It is widespread and is quite well known and many of you know it closely since what has made The New School, where we are gathered, one of the most valuable institutions of higher learning in this country is that it has valiantly housed –indeed it has been something of a hospice for– those suffering from an exclusion of unorthodox frameworks for thinking about a range of themes in a range of different disciplines.)

Dogmatism of this kind is also distinguishable from the third sort of flaw, obtuseness. To be dogmatic in this way is not at all to be lacking in the acuity that would recognize the force of counter-evidence and counter-argument. If one has failed to recognize any counter-evidence (downstream, in my metaphor), that is because one has (further upstream) not even so much as recognized the possibility of the framework from which it flows. It is not as if the counter -evidence is there for us to see downstream and we are not perceptive enough to see it. Rather it is not there for us to see downstream because we have not recognized the framework upstream, from within which it is visible. And this last failure is a kind of dogmatism, not stupidity.

Among disciplines, Economics provides the most gorgeous examples of this. It is perhaps the worst offender in inuring itself against alternative frameworks of thought and analysis. In fact, I will venture to say that I have never come across a discipline that combines as much extraordinary sophistication and high-powered intellect and intelligence with as much demonstrable falsehood. So, for instance, some of the most brilliant intellectuals I have known to this day make claims about the trickle down of wealth in capitalist economies and present them with the most sophisticated quantitative methods, despite the plain fact that wealth has not trickled down (at least not to the places where it needs to trickle down), anywhere in the world in the entire history of capitalist political economy. If a physicist were to make some of the claims that economists have made which have been falsified as repeatedly as they have, they would not only have their careers terminated, they would properly be the laughing stock of the profession. Now, there is no direct political influence that forces this sort of refusal to question, leave alone give up, one’s assumptions in a discipline such as Economics. The regulation is wholly within the discipline’s profession and even there, there may be very little browbeating or intellectual bullying, that is to say, very little explicit regulation. It is largely unconscious self-censorship –often done with career advancement in mind– that threatens academic freedom in such disciplines.

On the very evening after I wrote these words in a draft of this lecture, I was over at a dinner at my economist colleague Joe Stiglitz’s apartment and I impertinently told him that I was going to raise this point in a lecture I was to give at a conference on academic freedom at the New School the next day. His response was memorable. “Akeel, I agree with you about economists but I don’t understand why you are so puzzled. One would only be puzzled if one were making the wrong assumption about Economics. What you should be assuming is that –as it is done by most economists– Economics is really a religion. And so why should you be puzzled by the fact that they cling to and never give up their views despite their frequent falsification.” So I will rephrase my point: one apparently makes one’s way up in a church hierarchy by clinging fast to the orthodox faith.

But there is the following difference. The church has had a history of explicit and rigid regulation of what may or may not be said and pursued in its fold. But, as I said, if there is political (or corporate) influence in play in the sort of dogmatism I have described in Economics, it is not obviously visible and direct, and the protagonists in economic inquiry in universities would be quite genuinely clueless about it and, with no dishonesty, deny its influence. Sometimes, as in my own subject of analytic philosophy, where there is a great deal of exclusion of alternative frameworks for discussion, there is no political influence, however indirect, in play. If there is a question of power and politics involved it is entirely internal to the discipline, the power that is felt and enjoyed simply in keeping certain ways of thinking out of the orbit of discussion, forming small coteries of people referring to each other’s work with no concern that the issues they discuss are issues that have no bearing on anything of fundamental concern to any of a number of disciplines with which philosophy had always been concerned before, say, even fifty ago, and from which it has now managed to isolate itself for the most part. Richard Rorty had tried to raise Kuhnian questions for the discipline of analytic philosophy[vii] and spoken with eloquence about its insularity and he was certainly right to notice just how exclusionary the subject had become, the more it had become a profession in universities.

Moving away from specific disciplinary examples, the general point that emerges from these examples can be made if one recalls that De Tocqueville famously said “I know of no country where there is so little independence of mind and real freedom of discussion as in America.” And here is a wildly curious thing. At the same time, it is America that has more free institutions (including academic institutions) than anywhere else in the world. How can this extraordinarily paradoxical duality co-exist? What explains this paradox? I can’t possibly try to provide an explanation here,[viii] but whatever it is that explains it will provide a very good sense of the deep, that is to say, submerged, forms of academic unfreedom that exist in this country. When the freest academic institutions coexist with some of the highest levels of academic unfreedom in the democratic world, the sources of unfreedom are bound to be far subtler than is captured by the standard vocabulary of ‘suppression’, ‘brainwashing’, ‘political pressure’, ‘manipulation’, and so on. That is why this fourth phenomenon, this pervasive sort of dogmatism, is a far more interesting case of academic unfreedom than the third phenomenon in our taxonomy. When a person working with unorthodox frameworks of research is looked upon with perfect sincerity by professionals, as someone unfortunate and alienated and to be pitied as irrelevant rather than bullied and hounded, we know, not only that the political influences on these professionals are not even easily identified, leave alone easily confronted, we also know that this kind of thwarting of academic freedom needs a quite different descriptive vocabulary than I used in describing the third phenomenon.

Equally, I would insist that it is different from the first phenomenon of academic dishonesty as well because to accuse these professionals of dishonesty (rather than in their dogmatism unconsciously perpetrating academic unfreedom) would be to be glibly moralistic since (if I am right in making the upstream/downstream metaphor) it is not honesty that requires that people should be willing to allow frameworks of investigations other than their own. At any rate it is not honesty in the sense that is required to admit that one’s views have been refuted, when one has been shown evidence against them and one is not too obtuse to recognize the evidence. To insist that they are both a case of dishonesty would be to perpetrate a (not very good) pun.

The interest and subtlety of this exclusionary phenomenon, then, lies in its distinguishabilty from all of the other three in our quartet: academic dishonesty, academic stupidity, and straightforward and obvious forms of academic suppression. Despite its subtlety, it is a recognizable assault on academic freedom, and it is the more important to analyze in detail precisely because there is nothing as obvious (or infrequent) about it as there is about the efforts at external influence of Christian groups on science curricula or of Zionist groups on Middle East studies departments, in some universities. Being much more subtle it is also much more pervasive than these more obvious phenomena –and much harder to resist. Different people feel it differently at different times. Frameworks for serious research in race and gender felt it constantly for decades till as late as the seventies of the last century. Quite possibly more old-fashioned forms of humanistic scholarship in a range of literary disciplines began to feel it since the late eighties and nineties of that century. And I daresay, research programs which pursue seriously socialist forms of analysis feel it more than ever today in Economics departments…

I have tried in this lecture to shift attention away from the fallibilist epistemological presuppositions of metaphors such as ‘the market place of ideas’ and its classical Millian arguments for academic freedom, and I have tried to focus it instead on the need to diagnose the sorts of unconscious attitudes that make for unwitting disciplinary mandarins and gatekeepers, ‘normal scientists’ as Kuhn called them. One doesn’t need to be diffident in the conviction with which one holds one’s views in order to resist such attitudes. We should allow alternative frameworks not because we have some generalized doubt that we ourselves might be holding false views. We should allow alternative frameworks for quite different kind of reasons, also found in Mill’s writing on liberty, as I said, having to do with the fact that if we allow for frameworks of investigation other than our own, we make for an attractively diverse intellectual ethos and in doing so allow the creativity of different sorts of people and minds to flower. These sorts of consideration in favour of academic freedom, unlike Mill’s argument considered in Section II which appeals to all those capable of inductive reasoning and in pursuit of the truth, gives rise to a picture of academic freedom that appeals only to those who think that there is value in creativity and diversity in the academy. The appeal therefore is frankly disadvantaged by its less than universal reach. But, on the other hand, the picture can claim the advantage of not being landed with a bizarre conception of inquiry presupposed by Mill’s more ambitious argument and the metaphorical cliché that it has yielded about the ‘market place of ideas’.

The vital point I want to repeat in marking this difference from Holmes’ metaphor and from Mill’s meta-inductive argument on which that metaphor is based, is that if considerations about truth and falsity enter this picture, it is only, as I said, further downstream when something that other frameworks deliver might claim to be a truth that clashes with ours and provides some evidence or argument for us to give up some of our own convictions. But since those considerations do not surface upstream where we are pursuing the goal of inquiring into the truth in our investigations, that goal of pursuing the truth need never be conceived of as a goal whose success is necessarily opaque to its seekers, as in Mill’s argument for freedom. In our own pursuits towards the truth, we may be as confident in the truth of the deliverances of our investigations as is merited by the evidence in our possession, and we need feel no unnecessary urge to display balance in the classroom, if we have shown balance and scruple in our survey of the evidence on which our convictions are based, the only place where balance is relevant in the first place.

IV

Having said that, I should like to conclude with a point that rotates the angle a bit on the question of balance.

One of the questions that has most exercised scholars of academic freedom is the extent to which the concept and the policy applies to the utterances of a scholar not within the university but in what is called the ‘extra-mural’ context. Is a professor free to say things outside the university in public forums that would be unsuitable for one reason or another in the classroom or at official university events? There is a lot of interesting writing on this subject, some of the most interesting by scholars of the law. But I want to say something here that is a bit off that beaten track.

When it is not classroom curriculum and intellectual inquiry in the university but political debate in general outside the study and the classroom that is in question, there are good reasons why the views one expresses can and often should be substantially imbalanced. And by imbalance here, I don’t just mean that they should speak with conviction for one side of a disagreement, if that side has the preponderance of evidence on its side. That form of imbalance is what my critique of Mill and Holmes has tried to establish as perfectly appropriate in the classroom. But for extra-curricular and extra-mural public speech by academics, I have in mind the moral appropriateness of a further and more wilful kind of imbalance. To conclude in one’s thought what the evidence in one’s investigations dictate is not really a matter of choice or will.[ix] The evidence compels us, as it were. But to be imbalanced in the further way I am about to mention is a matter of will and moral decision. Let me explain.

I find it not only understandable but honourable, if someone speaking and writing in America finds it important to stress much more the wrongs of the American government and its allies and clients, like Israel, Saudi Arabia, Egypt, Pakistan, (now even India), Indonesia under Suharto, Chile under Pinochet…. rather than speak obsessively, as is so often done, about the wrongs done by Muslim terrorists or Islamic theocratic regimes or, for that matter, Cuba and North Korea… But if the same person was speaking or writing, say, in the Palestinian territories or in Arab newspapers, it would be far more admirable if he were to criticize Hamas or Islamic regimes like Iran’s….So also, unlike the many who were abusive towards him for not doing so, I find it entirely honourable that Sartre, living in Paris in the Cold War ethos, refused to spend his time criticizing the Soviet Union and instead criticized Western governments for the most part.

It is said that whenever Sakharov criticized the Soviet Union’s treatment of dissidents in the fifties, he was chastised by his government for showing an imbalance and not speaking out against the treatment of blacks in the American South. That is precisely the kind of imbalance that courageous academics are going to be accused of by the enemies of academic freedom in this country, and I hope that all of us will have the courage to continue being imbalanced in just this way.


[i] I am grateful to Jonathan Cole, Isaac Levi, and Carol Rovane, for helpful comments on a draft of this paper. I have also benefited from the discussion of a fragment of this paper presented to a workshop on pragmatism at the Institute of Public Knowledge at New York University, in particular the verbal commentary made during the discussion on that occasion by Craig Calhoun, Benjamin Lee, and Richard Sennett.

[ii] Actually, it was the idea and not the expression that Oliver Wendell Holmes put into the air in his dissenting opinion in Abrams Vs The United States. His own expression was “free trade in ideas”. The expression ‘market place of ideas’ was first used in the language of the Supreme Court in ‘Keyishan vs Board of Regents of the State University of New York’ (1967).

[iii] Hofstadter, R., Metzger, W.P. (1955). The Development of Academic Freedom in the United States (New York:

Columbia University Press, 1955, pp. 527).

[iv] John Stuart Mill, On Liberty, chapter II. (London: Longman, Roberts and Green, 1869.

[v] To be more precise and detailed, the pragmatist –Peirce, for instance, in his remarkably profound and original paper “The Fixation of Belief”, Collected Papers, volume 5 (Harvard University Press) —makes a distinction within inquiry between our settled beliefs and our hypotheses. See also Isaac Levi, The Enterprise of Knowledge (MIT Press 1986) for an elaborate and interesting deployment of this distinction to construct a theory of the dynamics of knowledge or what is sometimes called a theory of ‘belief revision’. Hypotheses don’t command our confidence in the same way that our beliefs do. Mill might be right to ask us, as inquirers, to have some diffidence in the way we hold our hypotheses to be true since, unlike a settled belief, an hypothesis, even by our own lights, is still up for grabs, it is not something we have decisively counted as having the full prestige of ‘truth’. But making this distinction and granting that he has a point about diffidence for one half of the distinction (hypotheses), does not help with Mill’s meta-inductive argument for liberty. The distinction merely says that unlike a settled belief (such as, say her belief that the earth is not flat), a scientist today might make an hypothesis that she is hoping to have confirmed by the evidence. She will hold the latter with diffidence but not the former. But, if the former is held without diffidence, then Mill’s argument for liberty does not hold for such settled beliefs. That is absurd, from Mill’s point of view. He would not have wanted ‘flat-earthers’ to be tyrannized and suppressed, so he would not have been prepared to restrict his argument for liberty to just hypotheses. He would have wanted freedom of speech and discussion to apply to the expression and discussion of all beliefs, settled and hypothetical. The trouble with his argument is that he extrapolates fallaciously from the diffidence that is properly advocated for hypotheses to all beliefs, even settled ones. And he must do so, since liberty presumably will apply to the expression of all beliefs. That is why I am suggesting that one should simply abandon this argument, with its fallibilist appeal to diffidence, altogether as providing the basis for free speech. It is the wrong basis for liberty. We should be looking for entirely different grounds and arguments for liberty, some of which Mill himself provides elsewhere in that work.

[vi] It might be thought that there is no very direct link between the broad liberal mentality towards freedom of speech and academic freedom that I am situating in Mill’s argument and this talk of ‘balance’ in the university. After all there are much more straightforward political motives for those interested in protecting Israel from the harsh criticisms of its actions towards Palestinians that it deserves, to insist on ‘balance’ in the way cotemporary Middle Eastern politics is taught. If both sides are constantly being presented equally, as is demanded by ‘balance’, then the force of such decisive criticism can be softened. I don’t deny that there are these political motives for demanding balance in cases of this kind as well as other cases. But we can’t forget that many political motives of this kind are cloaked in high-sounding intellectual arguments so that their nakedness, qua political motives, is hidden. Just think of the way slaves were said to be not quite ‘persons’ by ideologues rationalizing slave-ownership or the way natives were said to be lacking ‘rationality’ by colonists. These philosophical arguments are a constant factor in the pursuit of political motives. Millian forms of liberalism similarly often underlie (as rationalizations) the political motives for demands for balance in pedagogy.

[vii] Richard Rorty, Philosophy and the Mirror of Nature (Princeton: Princeton University Press, 1981)

[viii] Noam Chomsky and Edward Hermann in their rightly celebrated Manufacturing Consent have addressed this subject with the focus primarily on the media in this country and I am hoping that Jonathan Cole in his forthcoming magnum opus on the research university in America will –among the many other things about the American academy that that book is ambitiously intended to address—speak to this issue with the focus on the universities in this country.

[ix] I realize that it is a matter of will whether one presents in a classroom (or indeed in one’s research) what one has evidence for. That is why failing to do so is to be described as ‘dishonesty’. I am only saying here that when the evidence compels us to draw a conclusion, the will is not in play, it is not a matter of choice, even though coming to believe something on the basis of evidence is in the realm of the intentional.

Akeel Bilgrami is the Johnsonian Professor of Philosophy and the Director of the Heyman Center for the Humanities at Columbia University.

Like what you're reading? Don't keep it to yourself!
Share on Facebook
Facebook
Tweet about this on Twitter
Twitter
Share on Reddit
Reddit
Share on LinkedIn
Linkedin
Email this to someone
email