Some time back we puzzled about John Searle’s review of a recent book by Boghossian which contained an attack on a paper by Kathleen Lennon. Though Searle had not, as it turns out, actually read the Lennon piece, it was a target and springboard for his attack on everything other than a firm belief that one can have completely objective knowledge of a fully independent world. You are a hard core metaphysical realist or (gasp!) a relativist!
Lennon kindly came to our discussion and we suggested she reply to Searle. Her reply and Searle’s rejoiner are in the latest NY Review of Books.
Here are two telling passages from the exchange:
Lennon:
… my paper was instead addressing how rational assessment of knowledge claims is possible, if we accept the situatedness of knowledge seekers. It points out that feminists cannot be relativists for “feminist criticisms aimed to challenge and discredit the masculine accounts they critiqued, not simply to add a further perspective. This requires the possibility of rational encounters between the positions.”
One of the problems with Searle’s characterization of his supposed opponents is a running together of different positions. Those who argue that historical, social, and material locatedness constrain what we can discover and make sense of are accused of relativism … Yet those who argue that we are the source of the frameworks in terms of which we understand the world do not have to claim that we do this in a way unconstrained by an independent reality, even while accepting that such reality does not dictate to us the single best way of making sense of it.
Searle:
She says, correctly, that I had not read her article. I was reviewing Boghossian’s book, not her article. I have now read the article with some care, and I believe it contains a deep inconsistency. In her letter to me she denies that she is a relativist, and insists that the passages she quotes from her original article support her denial of relativism. But the key sentence in her original article is this: Theories cannot be assessed by reference to universal norms. This is an astounding claim, because it denies that there are universal norms such as truth, evidence, consistency, rationality, and coherence, by which we can assess theories.
In her original Aristotelian Society article (Suppl vol 71) Lennon looks at how the conditions creating one’s perspective may be invisible to one. Relatedly, it takes a particular cultural location to think that gender doesn’t matter, for example. It is interesting in this regard to see Searle as believing something quite similar; namely, that perspective does not matter, since we can hold the view from nowhere.
Feminist Philosophers 
It’s really sad when an eminent philosopher doesn’t know the difference between epistemology and ontology.
I think the best way to characterize Searle’s real problem (that is, the problem that’s bigger than quoting Lennon out of context) is as a version of the fallacy of equivocation.
Boghossian and Searle both seem to define universalism (about justification) as the view that all standards of justification are universal. So anyone who thinks that there are any standards of justification that vary with social context in any way is a relativist.
Now, so long as they’re clear and consistent, I’m happy to let them define the terms however they like. The problem, however, is that `relativism’ is now so broad that the universalism/relativism distinction is basically useless. `Relativism’, on this defintion, encompasses everyone from Peirce, Longino, and (IIRC) David Lewis, who think justification is mostly about being able to give good responses to good criticisms, to Evans-Pritchard, the anthropologist who famously argued that Zande witchcraft must be judged by its own standards of justification. Still less will you be able to defend such interesting claims as relativism implies idealism, or all relativism is of the `anything goes’ variety. Thus, when we have such arguments as
(1) Lennon is a relativist.
(2) Relativism implies idealism.
(3) Hence, Lennon is an idealist.
the `relativist’ in (1) doesn’t refer to all and only the people who believe in the relativism of (2).
This, by Lennon, struck me as the most telling passage, though of what I would like to know:
“[T]hose who argue that we are the source of the frameworks in terms of which we understand the world do not have to claim that we do this in a way unconstrained by an independent reality, even while accepting that such reality does not dictate to us the single best way of making sense of it.”
Is this some kind of sweeping transcendental idealism, or a claim restricted to, as Searle puts it, “where human reality is concerned”?
I think it means quite simply that we learn what the norms of rationality are over time, and because we are still learning them, no one can claim to know them all already yet, as Searle seems to.
I think the example of the ruler is an interesting one. Certainly, humanity didn’t emerge from apehood with the ability to tell how straight or long things are. We had to fashion first primitive rulers, then use those primitive rulers to make more accurate ones. But it would have been absurd for a Paleolithic Searle to say, “How can you want to use that new feminist ruler? The old masculinist rulers clearly shows that it is crooked!” There’s no existing ruler that we can appeal to in order to decide which ruler is the straighter or the better at measuring. But we can still try to decide these things through cultural evolution and through interaction with the mind-independent world. If we’re lucky, we come up with better normative standards over time.
What’s particularly ironic is that this same process is blatantly obvious in that most treasured jewel in the crown of analytic rationality—deductive logic. No analytic worth his/her salt would ever use Aristotelian logic for the coup de grace of a paper. Nor can they choose to use modal logic S3 instead of S5 or whatever without some remarks about the assumptions involved in using one or the other. The standards of what constitutes doing good logic have evolved over time, but they evolved through humans thinking about them, debating them, and using them. The same as the ruler, the same as any other human artifice.
In that sense, unless one believes the human mind has infallible access to the transcendent use of reason, then one has to believe that the debate about what is and isn’t good reasoning is still open and on-going. Which means we need to be aware of how our social locations give us sets of norms, and these norms may or may not be the right ones. In order to get better norms, the feminists prescribe diversity and openness to other viewpoints. This is also a socially situated norm. Is it a good one? We can only find out by using it and hermeneutically refining it.
Biological classificatory terms have a number of nice examples of fairly foundational notions that don’t cut reality at the joints, and indeed leave one thinking reality doesn’t have those joints. Just how to determine biological kinds is not just debated, but it’s hard to see how one can get more than “this is the best classification we have found for our purposes.”
As criteria for kinds shifts, so membership in kinds can shift. Pandas got thrown out of the bear category, but now they’re back in it.
Woah, pandas are bears? I did not know that. Koalas still aren’t bears, right?
I didn’t know that pandas got thrown out of the bear category! I thought it was just red pandas – apparently unrelated to the big bamboo eater – that weren’t bears. I will feed that info into the ongoing are pandas bears argument my friends keep having on drunken nights. (On a par with the ‘are dishwashers more environmentally friendly than washing up by hand?’ argument.)
Here’s the wiki take on the controversy, with dates and all:
Thanks! Really interesting.
I took Searle to be, in an unnecessarily combative way, to be getting at something like the distinction made in philosophy of science between the “context of discover” and the “context of justification”, and saying that it’s a confusion to move directly from the fact that social factors are extremely important to the context of discovery that they are important in (or undermine) the context of justification. That claim is, I think, right. Now, this doesn’t mean that issues from the context of discovery are never important in the context of justification, just that you can’t move from one to the other without more argument than is often given. If this is what Searle is saying, then I think his general point is right, even if he’s making the cuts in the wrong places sometimes. Does it seem to others that he’s saying something very different from this?
That, Matt, is how I read Searle. Yet Lennon (and, perhaps, Carl), as quoted on both page 6 of Boghossian’s book and in Searle’s reply within the exchange, seems to be claiming that context of discovery is always important in the context of justification (“all knowledge is situated knowledge…”).
This I simply don’t understand. How does context of discovery bear on the context of justification in, say, arithmetic and basic logic? It seems like immemorially familiar self-reference issues immediately bob up, since any attempt to flesh out such a view is, in the very process of articulating it, going to have recourse to the universal normative authority of precisely what it’s purporting to relativise, no?
How does context of discovery bear on the context of justification in, say, arithmetic and basic logic?
In basic logic, a person’s views about validity are going to be affected by the logical theory they subscribe to antecedently, and possibly by the natural language they speak as well. Does “not P or Q” entail “if P, then Q”? Depends on whether the conditional is the material conditional, or the Lewisian counterfactual with strong centering, or what. Maybe there is a universal logic into which all logics can be translated, in which case the move to relativism is too quick. But do we know that’s the case? I don’t see how to compile an exhaustive list of possible logics.
I’m less sure about arithmetic. If one takes intuitionist arithmetic seriously as a rival to classical arithmetic, then one might think that there are at least two incompatible yet somehow equally legitimate conceptions of what the truths of arithmetic are. I’ve been reading up on intuitionism, and cen’t decide whether to take it seriously. (So it’s not the case that either I take it seriously or I don’t…)
There’s also the issue of category-theoretic vs. set-theoretic foundations. I don’t completely understand the relationship there.
Sorry, that should be something like “blamelessly affected” rather than just “affected”, so that I’m answering Rob’s question.
Rachel said: In basic logic, a person’s views about validity are going to be affected by the logical theory they subscribe to antecedently, and possibly by the natural language they speak as well.
Sure. That’s all context of discovery, though, the same way that, say, Aristotle couldn’t have thought up Frege’s logic because various related ideas (about algebra, maybe) hadn’t been developed yet. But that’s irrelevant to the question of which is a “better” logic for some purpose. Similarly, the fact that Kepler had all sorts of crazy ideas about what he was doing, ones we now think of as pseudo-scientific, doesn’t in any way impact the truth of his laws. These are simple cases, and in ones where we deal with what Hacking calls “reflexive kinds” (or something like that- I’m too lazy to look it up now) the issue can get really complicated, but I think the basic point has to be right, and I think it’s something like that that Searle was getting at. (I’m not certain about this, though- I only skimmed the original review and didn’t even read the exchange with a super close eye- I just read it while drinking coffee at a book store a few days ago.)
(I had an earlier comment on this thread. Was it caught in the spam filters?)
Matt and Rob — You’re a bit behind the times on the context of discovery/context of justification distinction, at least with respect to philosophy of science. Since the ’60s, historians and sociologists have argued that the processes of formulating a scientific theory can’t really be teased apart from the processes of giving reasons in support of the theory. Philosophers of science have by and large conceded this point. Contemporary debates have to do with (a) exactly which processes count as giving reasons and (b) whether (and, if so, how) ethical and political values play a role in these processes. I suppose one could try to read (a) in terms of the discovery/justification distinction, but it would be an awkward fit.
Consider the development of a model of some particular aspects of climate change. An initial model is constructed, based on some antecedently-accepted physical theories, and found to work well with some data, but not others. The model is adjusted in various ways, improving its fit to the data. A paper is published, and the authors’ work is subject to criticism in the peer review processes and some followup emails. Some of these criticisms encourage the authors to adjust their model in certain ways; other criticisms encourage adjustment in other, incompatible ways. The authors prioritize these criticisms, make certain adjustments to their model (but obviously not all the adjustments any one critic wanted), and write a follow-up paper. Lather, rinse, repeat.
Now, what parts of this story describe the context of discovery, and what parts the context of justification? Note that, as Rachael suggests, you can tell exactly the same sort of story about the development of theories in mathematics and logic.
Surely the discovery/justification distinction is questionable. Naturalistic epistemology is in part motivated just by the fact that our justification often consists in how we come to believe and perhaps the context. We don’t have little justificatory arguments for much of which we believe, as Hume well say and much in cognitve science confirms.
I also don’t get why maths is supposed to be the model here. Searle’s example are some of the cliches of ordinary or scientific contexts. It seems to me to lead here to a conflation between truth and justification.
Noumena, I think we were writing at the same time.
Noumena and JJ- In fact I read lots of philosophy of science stuff. It’s not my main area of work, but it’s a major interest of mine and lots of top people working in the area push a line much harder than the one I’m pushing, so it won’t really do to just suggest I read more. Let’s look at the climate change story. Nothing you say goes against anything I say (or what Searle says, as far I can understand him- again, I’m not really sure what his position is.) In the end, either the climate is getting warmer or it isn’t The fact that lots of political (in a broad sense of the term) elements are important in how we come to believe this is interesting and important, but of course that doesn’t make any difference to whether the climate is getting warmer or not. So, I don’t really see how this supports the idea that there’s not a distinction to be made between context of discovery and of justification.
JJ- but in naturalized epistemology, isn’t the question (or one of the questions- there are many, obviously) things like, “is the process by which we come to believe X a reliable one?” But to answer that, you have to think there’s a standard by which reliability can be measured. And that’s just the sort of thing (I think) Searle is getting at.
Finally, note that there’s no need at all to be a metaphysical realists to make this distinction or think it’s important. It’s made and used (though not always put in these terms) by people like John Dupre, Arthur Fine, Michael Friedman, and so on. Even standpoint epistemology of the sort found in Marx and (in an unfortunately worded way) Katherine MacKinnon, has to have some version of this distinction to work at all.
Matt: naturalized epistemology doesn’t have to be reliability. And when it does go reliabilist, the conception of realiability does not necessarily commit one to Searle’s realism.
Matt, I don’t understand your response. I think you’re confusing relativism about justification — an epistemological view — with some sort of anti-realism — a metaphysical view. Discovery/justification was a distinction made to strictly demarcate epistemology from history and sociology, and the concession philosophers of science have made since Reichenbach’s day is that this strict demarcation doesn’t work. There’s nothing anti-realist running around in there.
In my earlier post — the one that, sadly, was eaten by the spam filter — I argued that Searle and Boghossian were equivocating on `relativism’, and using this equivocation to argue for such blatantly false claims as that all relativists are anti-realists. I’m worried that you’re making the same mistake, especially since you mention John Dupre; Dupre is both a realist and an opponent of value-free conceptions of objectivity (hence, is a relativist about justification).
“In the end, either the climate is getting warmer or it isn’t.”
If we fix the meaning of “warmer” in advance, yes. Otherwise, the term is vague and bivalence won’t necessarily hold. But fixing the meaning of “warmer” is going to have to be a political act, because how else we decide if the heat-island effect in cities counts or not, or whether we want global mean temperatures or global mode temperatures?
It’s dangerous to think about science in terms of propositions, because science is conducted in natural language (albeit technical natural language), and natural language doesn’t have all of the nice properties of propositions like bivalence and so on.
I can’t claim to fully understand Carl’s post immediately above, so it may well simply be a function of ignorance, but it sure does sound to me like an example of just what Searle and Boghossian are criticizing… And the reason I brought up arithmetic or basic logic earlier is because of Lennon’s claim that “ALL knowledge is situated knowledge” (my emphasis). I still can’t fathom how, for instance, the knowledge that 2+3=5 has anything significant to do with the situatedness of the one possessed of it. The same, to follow Thomas Nagel (in The Last Word, Oxford, 1997), would seem to apply to something like modus ponens. In these instances, belief is not, I would think, an ungrounded *assumption* ineliminably affected by our situatedness, but *acceptance* of unqualified, universal validity.
I mean, above, a function of MY ignorance, not Carl’s!
You know about the sentence/proposition distinction, right? The sentence “the cat is on the mat” may be true or false depending on what cat and what mat we’re talking about at particular time, but once we abstract the context, the sentence is supposed to correspond to a proposition OnMat(cat) which is eternally true or eternally false.
The sentence “two and three make five” corresponds to the eternal truth of the proposition “2 + 3 = 5” but surely this truth was discovered by someone, probably a Babylonian although I don’t know the history of proto-math well enough to say. Now suppose the context we’re talking about is a 2 bit computer processor. In that case “2 + 3 = 1″* because the bits overflow the largest possible number representable on the system. (Of course in the real world no one uses a 2 bit processor, but overflow is a real problem that computer programmers have to take measures to deal with. Typically the problem is on either side of 32768.)
“2 + 3 = 1″* doesn’t affect the eternal truth of “2 + 3 = 5” because both of those things are just as eternally true within their own contexts. Think also of all of those old philosophy books with the example of a triangle with 180 degrees of internal angles. That’s still eternally true… in a Euclidean geometry. The old philosophers thought that would be always true anywhere and without context. It turns out that it’s always true if you can specify ahead of time, “I mean in a Euclidean space.”
Even our beloved modus ponens only sort of works in paraconsistent logic. To make it work, you have to have a difference between “~A v B” and “A -> B” with the latter being a stronger claim. So even logical claims have to be justified relative to a specific system of logic.
The trouble for human beings is that we always start out on the sentence side of things and have to hope we’re getting to the proposition side. For math getting to proposition side isn’t too hard, because the terms are mostly empty (what is 2 anyway?) but for science it can be quite difficult. Talking about global warming involves a lot of terms that are only vaguely defined. It could involve global mean temperatures, global median temperatures, global mode temperatures, temperatures just on land, temperatures just in cities, number of record breaking high or low days, how high the highest high is, etc., etc. Accordingly we can’t appeal to the law of excluded middle until we’ve figure out what we’re talking about first, and deciding what words mean is a human process.
Yikes, you all! A now quite well entrenched theory has something like 3+2=5 as close to something we just sense. We don’t discover it any more than we discover that objects are colored (or coloured). Anything much more than that and our basic sense of numbers gets fuzzy and we need some sort of system or code. In fact, lots of animal species can grasp that 2+2 = 4. (This is very distinguished work by Dehaene (and Spelke)).
I think it is important that Lennon is talking about things we believe about an independent world. I’d be inclined to restrict my discussion of it to concepts and beliefs that are more empirical, since it’s really hard to figure out what numbers are. Much harder than what atoms are, or are supposed to be. IMHO.
I can’t think of any good reasons for rejecting the claim that 2+3 = 5, but Vann McGee has some persuasive arguments that modus ponens is invalid.
I’m not certain that I understand how to interpret Lennon’s rejection of the claim that “theories can be assessed by reference to universal norms” in the context of logic. (Note: Italicized phrase inserted in light of comments below.) Certainly, it entails that each logical theory can be assessed by reference to universal norms. I was assuming that it also entailed that if there were such universal norms, that they would be logical norms, and hence that there would be a single, correct way of doing logic. Anybody who did logic not in the correct way would be making a logical mistake. But then, at most one of the following views would not be based on a logical mistake: classical logic, intuitionism, supervaluationism, the “logic of paradox” LP. Either that, or the proponents of those views would not genuinely be disagreeing. I think that the disagreements are both genuine and blameless. (Matt’s reply to me shows that I need the blamelessness assumption, and this might be where he disagrees with me.) So logic, it seems, cannot be assessed by reference to universal norms.
Well… the above proof was a reductio, and a multiple-conclusion reductio at that. If it’s controversial whether reductio is valid, then it’s controversial whether I’m entitled to the conclusion. But at the very least, if I accept the validity of the argument from the disputed claim (that logical theories can be assessed by reference to universal norms), I’d better not assert the disputed claim, whether or not I get to assert its negation.
I think most people who believe in a single, correct logic also think that anyone who does not accept that logic is making a logical mistake. But I’m curious to hear from Matt and Rob.
I was going to complain that Matt was conflating justification and truth, but if one is a logical pluralist, and one thinks that validity is truth preservation, then maybe one has to be some kind of truth pluralist as well.
jj, I think you’re right about the intended scope of Lennon’s original claim, but I think the claim is also interesting and defensible in the case of logic. Let me know if this gets too distracting; I’m really enjoying the discussion but it’s probably no longer germane to the original point.
RB, no please, do continue. I meant it as a way of defending her claims, but I didn’t intend to limit the discussion.
I’m having trouble with your second para above. Normally one can tell when a “not” is left out, but not in this discussion. Lennon does have a “not” in the sentence you quote.
Excellent; thanks! I’m not sure what caused the misattribution in the beginning of the second paragraph; better emend as follows.
I’m not certain that I understand how to interpret the claim that Lennon is rejecting–that “theories can be assessed by reference to universal norms”–in the context of logic.
I think everything else is OK as it stands.
I’ll change it…