So I was brought up in a culture of conviction and therefore in some ways my upbringing was like that of those raised in religious belief. I shall set out the early stages of my political and religious and irreligious beliefs in Lecture 2. But before I describe the development of my convictions I should like to explore some aspects of the development of conviction in general aspects that I came to find puzzling as I reflected on my own development. In the case of what are properly called convictions but even as we shall see in the case of beliefs which might be considered too cool in temperature to be called “convictions” there is a problem about how we manage to go on believing what we were raised to believe in the face of our knowledge that we believe it because (in a certain sense which I shall specify at p. 10 below) we were raised to believe it. F. Scott Fitzgerald's point about the power of the past (see p. 1 above) is comparatively easy to accept with respect to our feelings and emotions. But for many of us his point also has force with respect to our beliefs and that is more unsettling since it raises questions about the rationality of those beliefs—questions that are more unsettling than parallel ones that might be raised about our feelings and emotions.

Suppose that identical twins are separated at birth. Twenty years later they meet. One was raised as and remains a devout Presbyterian. The other was raised as and remains a devout Roman Catholic. They argue against each other's views but they've heard those arguments before they've learned how to reply to them and their opposed convictions consequently remain firm.

Then each of them realizes that had she been brought up where her sister was and vice versa then it is overwhelmingly likely that (as one of them expresses the realization) she would now be Roman Catholic and her sister would now be Presbyterian. That realization might and I think should make it more difficult for the sisters to sustain their opposed religious convictions. Or to come closer to home—or at any rate to where I am—suppose I were to discover that I have an identical twin who was raised not in a communist home but in a politically middle-of-the-road home and that my twin has the easy tolerance toward limited inequality which I learned to lack. That I confess would disturb my confidence in my own uncompromising egalitarianism and not because my twin could supply me with an argument against egalitarianism of which I was previously unaware.

To be sure the surprise which the twin sisters undergo need not make it hard for them to remain Christians; and the revelation which my twin brother and I experience need not make it hard for us to remain anti-Tory. But such further difficulty will supervene if the sisters turn out to be not twins but triplets and they now meet their long-lost and now Jewish third sister. Which so far as it goes will no doubt leave their belief in (a) God secure until they meet their long-lost and now atheist quadruplet whose confidence in her atheism may be shaken when she confronts them. And a similar extension of the story about me and my brother can of course also be rolled out.

Now most of us are solo-birth children; most of us were that is born twinless tripletless and so forth. But it would be crazy to infer that the story about the twins has no bearing on our convictions. That I am in fact twinless should not reduce the challenge to my inherited convictions which is posed by the story I've told. An entirely plausible story could be told about a hypothetical disagreeing twin and it would or should be just as challenging as a true story to those of us who believe what we were brought up to believe.

That is not of course all of us. But it is very many of us. And although it does not follow from the fact that we believe what we were brought up to believe that we believe it in any sense because we were brought up to believe it it is very widely true that people do believe what they do in some sense¹ because they were brought up to believe it: the statistics of parent-to-child belief replication prove that. And I think these stories about twins and so forth should give pause to those of us who are steadfastly devoted to the beliefs of our upbringing while being aware that people of different upbringing are steadfastly devoted to beliefs contrary to our own. It should give us pause that we would not have beliefs that are central to our lives—beliefs for example about important matters of politics and religion—if we had not been brought up as we in fact were. It is an accident of birth and upbringing that we have them rather than beliefs sharply rival to them and (here's the rub) we shall frequently² have to admit if we are reflective and honest that we consequently do not believe as we do because our grounds for our beliefs are superior to those which others have for their rival beliefs.

The problem I am posing does not require a narrow view of the sorts of grounds that one can have for a belief. Consider for example the person who says that she believes in God because she underwent a profound experience that she cannot fully describe and experience that induced faith in God within her. I am not skeptical about that claim as such I do not find appeal to a special religious experience intrinsically unacceptable. The skepticism in focus here arises rather when we compare A's grounds for believing that p with B's grounds for believing that q and we notice that however good or bad those grounds may otherwise be they do not relevantly differ in quality; so that so it seems it should be difficult for each to maintain his convictions when he confronts the other. For neither can reasonably believe that he believes what he does rather than what the other does because he has better grounds for his belief than the other does for his as opposed to: because he was brought up differently.

So I do not say: because faith-inducing experiences are not pieces of scientific reasoning they cannot credentialize belief. What I do say is that if for example profoundly Catholic religious experiences tend to take place in Catholic homes and profoundly Protestant ones in Protestant homes then it looks as though both Catholics and Protestants should be wary about the messages apparently conveyed by their religious experiences.

Let me say something about what I mean by “because” here in such sentences as “she believes it because she was brought up to believe it.” When I say such a thing here I do not mean that her belief is groundless. Nor do I mean to deny that she has reflected on and assessed the grounds she has for holding it and continues to hold it only because her belief survived that reflection. I have in view throughout nurtured beliefs which have indeed passed the test of reflection for the believer. But even so even though the beliefs I am targeting are not (any longer) held for no reason they are there in a crucial sense because of the believer's upbringing. The reflective nurtured Catholic and the reflective nurtured Presbyterian may for all that I am concerned to contend believe what their beliefs have in common entirely because they have drawn the right conclusions from the evidence available to them. But the whole explanation of the difference between their beliefs the explanation for why one believes p as opposed to q and the other believes q as opposed to p lies typically in their upbringing³ rather than in the quality of the data that were presented to them or in the quality of their reflection upon that data.

I emphasize that contrast because what disturbs me in the cases under inspection here is not just in itself that the person believes differently because of a different upbringing but that she cannot honestly identify a relevant further difference. These cases differ from that for example where one person was brought up to believe that the earth is round and another that the earth is flat. When the round-earther reflects that had he been brought up flat-earthly he would now believe the earth to be flat that need not give him pause for he can reasonably say that his grounds for believing it to be round are overwhelming. Round-earthers (justifiably) think they can prove their position. But Catholics—or at any rate enough Catholics to make my point interesting—would acknowledge that they are in the same epistemic boat as Presbyterians are that the Presbyterians’ grounds are no worse than theirs are. Since in the relevant cases you can't find anything except nurture that makes the difference since you can't say on independent grounds that their nurture was defective you can't say that you have better grounds for believing that p than they have for believing that not-p. And this flies in the face of what seems to me an undeniable principle about reasons for belief—namely that you lack good reason to believe p rather than a rival proposition q when you cannot justifiably believe that your grounds for believing p are better than another's are for believing q (call that principle “the Principle”).⁴ For you have to believe that your grounds make it more likely then not that p is true and they don't do that if they make p no more likely than his grounds make q likely.

Now paradox looms here not because there exist the truths about cultured beliefs that I have labored but because we the believers or anyway those of us who are reflective are (at least implicitly) aware of these truths. Thus for example no intelligent and reflective Scots Presbyterian can herself suppose it irrelevant to the explanation for why she is a Christian and not a Jewess that she did not wind up in the wrong hospital cot at birth and her reflection may tell her that she would not then have received a less good case for Judaism than the one she actually got for Christianity. And while you may think it unsurprising at first glance that I hold the egalitarian views that were instilled in me you should perhaps find it a little surprising that I do when you realize that I stick to them even though I know that I hold them because they were instilled in me and that less radical views with no less good epistemic credentials might have been instilled in me if for example I had been brought up in the upper-middle-class Jewish part of Montreal instead of in the working-class Jewish part of Montreal.

To believe that p rationally⁵ is to believe that one has a good reason to believe that p that one has grounds for believing that p which constitute a good reason for believing it;⁶ and in particular that the grounds for holding the beliefs one does must be such that they give one good reason to hold those beliefs as opposed to the competing beliefs that others hold. To be sure one need not believe on pain of irrationality that one could state the grounds for one's belief forthwith or even that one could with sufficient time recover them. One can believe that one has forgotten irrecoverably what one's good grounds are. Or one can believe that they are there in one's mind but not yet capable of being articulated. One can say that one senses that one has good grounds for believing that p which have not yet surfaced into consciousness. But so soon as one confesses that one's belief lacks appropriate grounding one condemns oneself as irrational.⁷

Those qualifications constitute a partial explication of the claim that to believe non-irrationally that p is to believe that one has good grounds for believing that p (grounds that is which constitute a good reason for believing that p). But the qualifications do not erase the paradox toward which I am moving. For even when we review the qualifications it will remain evident in leading cases of nurtured disagreement that what distinguishes me from her is not that I possess special grounds of a kind that she lacks or that I have a hunch of a kind that she cannot claim to have but just my upbringing. And then I appear to be in difficulty. For the fact that I was brought up to believe it is no reason for believing it and I know that.⁸

We have to believe about our beliefs that we have good reasons for holding them. Yet even when we become apprised of these facts about the genesis of our convictions and these norms internal to the holding of conviction we or many of us still don't give up our beliefs; we feel that we needn't in the face of all that give them up. So it seems we can prove what we think we know is false: that we should give up our (controversial) inherited beliefs. An exceedingly familiar fact that belongs to what can be called the elementary sociology of conviction one that we all know about from our ordinary experience thereby appears to generate a paradox.

The argument implicit in the foregoing discussion the argument which is the locus of paradox has three premises. The premises look hard to fault and the conclusion seems to follow; yet the conclusion looks hard to accept because at least many people's considered convictions contradict it and their behavior appears to conflict with it:

Call that argument “the Argument.” Note that its conclusion flatly contradicts what we (or anyway many of us) are confident is true for we do not usually think that nurtured beliefs of the sort under contemplation here are irrational. That is the interest of the argument; that is what gives it its air of paradox whether or not it is sound.

And I am indeed not sure that the argument is sound. If the Argument is unsound and (4) has therefore not been shown to be true then that would be good news but we would then have the intellectual difficulty that it isn't so easy to see what's wrong with the Argument. But if the Argument is indeed sound then our intellectual difficulty is not of course to show what's wrong with it but to determine how (some of us) manage to sustain our strong impression that the beliefs here in view are not irrational. If the Argument is sound then people are starkly irrational in contexts where we do not normally account them irrational. It is not of course news that people can be irrational nor is it news that it can be puzzling to see how they manage to be so. What's puzzling here if the Argument is sound is not that people can be irrational (that puzzle especially strongly raised in this sort of case) but just that we do not normally consider beliefs of the sort that I have identified as instances of our (perhaps even commonplace) irrationality.

Now there are many significant challenges to the Argument and examining all of them would take us too far afield.¹¹ But I shall now address three of them the first two of which so it seems to me are defeated by the self-same compelling counterexample which is presented in section 3.

The depth solution works well enough for an easier question than ours—the question namely which asks: How can equally intelligent and open-minded people have utterly opposed beliefs? But that is not our question because it formulates a purely third-person problem. The depth solution collapses just where we need it—at the reflexive level where the relevant questions are posed in the first person. For how can I stick to p even when I can truly say that I see a deeper case for it than I do for q when I have no reason to think that the q-believer sees a case for q that is less deep than the case I see for p? The strength of that question is confirmed by the example laid out in section 3 below.

The second challenge is not to the soundness of the argument but to the interest of its conclusion. It runs as follows: “It is no accident that the beliefs which come from nurture and more generally beliefs which display an irreducible element of (comparative) groundlessness are characteristically religious or heavily political in subject matter. The strength with which such beliefs are held the emotion that characteristically attaches to them—all that makes them suspect anyway; and it is therefore not surprising that their genesis too is suspect. Rational people should abandon them. They partake more or less of fanaticism.” Call that the credal cleansing proposal.

There are two objections to the credal cleansing proposal. I am more sure of the soundness of the second one but I enter the first one too because it would be extremely interesting if it is sound and I am not sure that it is not.

The first objection is that if we were to abandon all religious and heavily political views then most of us would be stripped of the convictions which structure our personality and behavior. Life would be bland lacking in élan and direction. Everybody would be l'homme moyen sensuel. Maybe irrationality is preferable to a dull existence.¹²

And the second objection to the credal cleansing proposal is that it is to an important extent false that beliefs which have a significant trace of nurture are all of a religious heavily political or similarly high-temperature sort. As I shall show in the next and closing section of this lecture quite reconditely theoretical beliefs also display substantial traces of nurture. It accordingly begins to look as though the credal cleansing policy is more drastic than it presents itself as being. If their suspect genesis shows that religious and political beliefs have to go then much more has to go with them.

The credal cleansing proposal is also very drastic because it would cut away a great mass of nontheoretical quotidian beliefs. For a belief's being due to upbringing is neither necessary nor sufficient for it to be at variance with rationality; I have focused on upbringing because it is nevertheless an especially potent source of beliefs that have the power to resist rationality. Plenty of other widespread belief differences are not more rationally based than inherited belief differences are but reference to nurture immediately generates a host of compelling examples which are especially relevant here because of my focus on my own beliefs in the next lecture.

Realizing that the Argument would also impugn casual beliefs of ordinary life one critic of the Argument (Brian Barry) adduced the following example which is the third challenge to the Argument that I shall consider. Your and I go to the same play and we disagree about how good it is. There must be something about us that accounts for the disagreement something about personality or history or whatever since we are responding to the same thing. We may agree about the good-making and bad-making characteristics of the play yet simply be differently impressed by their relative importance. Does this make it irrational to hold the views we do? Why should it? We both have reasons for our views and we admit they're not conclusive. If though our disagreement is not irrational then the standards that I have laid down for rationality are pitched too high.

There is in Barry's example something unapparent that explains the difference between our judgments about the play. Barry says that we are not being irrational but I think that whether or not we are being irrational here depends on our view speculative though it perforce is about what the unapparent explanation of our difference of judgment is. Each of us might think that his faculty of judgment and/or sensibility is in general superior to the other's and that this explains why we do not agree. If so we can indeed comfortably persist in our disagreement but the case is then not parallel to one where we assign our difference to mere differences of nurture. On this interpretation of Barry's example it does not represent a counterexample to my position.

If however we avow that the reason we're differently impressed by the play is not different quality of judgment and/or sensibility but simply different background etc. then our disagreement on my view becomes peculiar—not that it breaks out but that we persist in it at that point. It is at that point that we become placed as nurtured believers are; but then the suspicion of irrationality persists and the example is once again no special threat to my claims.

I do not accordingly regard Barry's example taken in whichever of the two ways it may be interpreted as an embarrassment for my view that persistence in a belief which one assigns to one's upbringing appears to be irrational.

In 1961 I was on the eve of doing graduate work in philosophy and I was able to choose between going to Harvard and going to Oxford. Against the advice of some of my McGill University teachers (and of one in particular who said I recall that Harvard's Willard Van Orman Quine “could put [Oxford's] A. J. Ayer in his pocket”) I chose Oxford not because despite the stated advice I was more drawn to Ayer than to Quine—I was agnostic about their comparative merits—but because it seemed much more exciting to leave Montreal for Europe than to leave Montreal for Massachussetts.

I was pretty ignorant of Oxford-style philosophy before I arrived in Oxford and I spent my first year there absorbing what I could. One thing I learned to do was to ask questions about the status of truths and of supposed truths. If someone said “p” you'd then pounce as follows: Is that analytic (that is true by virtue merely of the meanings of the words in which it is expressed) or is it synthetic (that is true for some more substantial reason)? An example would be the contrast between “All bachelors are unmarried” which is analytic if anything is and “All bachelors are tetchy and demanding” which may be just as true as that they are all unmarried but whose truth depends on more than just that they are rightly called “bachelors.”

“Is that analytic or synthetic?” was a terrifically important question in the Oxford of 1961. If as sometimes happened someone perhaps from Germany or Italy said something rather grand and general such as that memory falsifies experience or that God is everywhere or that reason is tripartite or that man is distant from being or from the isness of being then he or she would be subjected to the cited interrogative pounce and if the response was “Um … er …” as it often was then that would be that as far as they were concerned.¹³

By the end of my first year at Oxford I was reasonably agile at distinguishing (apparently) analytic points from synthetic ones and I enjoyed doing it. Then in the autumn of my second Oxford year I read a famous article by Quine called “Two Dogmas of Empiricism.” Quine there said among other things that it was a false dogma that truths could be sorted into analytic and synthetic ones for there was no such thing as analytic truth all truths depended for their truth on the way the world is.

I did not want to believe that Quine was right but I also did not want to believe or disbelieve anything just because I wanted to believe or disbelieve it. So I worked hard at Quine's arguments and in the event I decided that they were not good arguments. In reaching that conclusion I was helped by my reading of various anti-Quine articles—by in particular H. P. Grice and P. F. Strawson and Jonathan Bennett and also by a pro-Quine-ish (or pro-ish-Quine) article or two by Hilary Putnam. I still think Quine was wrong but I don't care about the issue as much as I did then.

Now people of my generation who studied philosophy at Harvard rather than at Oxford for the most part reject the analytic/synthetic distinction. And I can't believe that this is an accident. That is I can't believe that Harvard just happened to be a place where both its leading thinker rejected that distinction and its graduate students for independent reasons—merely for example in the independent light of reason itself—also came to reject it. And vice versa of course for Oxford. I believe rather that in each case students were especially impressed by the reasons respectively for and against believing in the distinction because in each case the reasons came with all the added persuasiveness of personal presentation personal relationship and so forth.¹⁴

So in some sense of “because” and in some sense of “Oxford” I think I can say that I believe in the analytic/synthetic distinction because I studied at Oxford. And that is disturbing. For the fact that I studied at Oxford is no reason for thinking that the distinction is sound. Accordingly if I believe it sound because I studied at Oxford if that explains why I believe in it whereas say Gilbert Harman does not then my belief in the distinction is ill-grounded. But I can't comfortably believe that a belief of mine is ill-grounded. So I can't comfortably believe that the only reason I believe the distinction to be sound is that I went to Oxford. I can readily believe that this is how I came to believe it in the first place but I have to believe that my present reason for sticking to it is that I have good reasons to do so. But what in the world are they? Might they include the reason that Oxford is a better belief-producer than Harvard is because for example it has better architecture? (And one thing that I must believe of course if I stick to my belief in the analytic/synthetic distinction is that I am lucky with respect to the view I have of this matter that I went to Oxford rather than to Harvard. Maybe I am lucky that I did anyway.)

Consider in the light of the analytic/synthetic distinction example the depth solution which I floated and sank in section 2. Perhaps a deeper case for the distinction was available at Oxford than at Harvard. But that doesn't make it right for me to believe the Oxford doctrine now since I can be pretty sure that a commensurately deeper case for rejecting the distinction was available at Harvard. You see the deeper case at home but that doesn't help since you know that the other guy sees the deeper case against it where he is.¹⁵

Well enough of these morose meanderings. In the next lecture I shall by before you some of the history of my own formation of conviction under the title “Politics and Religion in a Montreal Communist Jewish Childhood.”