forum 2: week of 16 Jan - Lewis

Lewis is trying to reconcile skepticism and fallibilism. The result is contextualism: what someone can be truly said to know in one conversational context is different from what they can be truly said to know in a different context. Two things are worth noting
- this is not just that standards of evidence or confidence are higher in some contexts: it goes deeper (read carefully to see how)
- the way a person's evidence can rule out a possibility is rather unusual. Try to get a grasp of what this amounts to.
Now some questions for you to react to.
~ Lewis gives a number of "rules" for what possibilities can be ignored (so they don't have to be eliminated). Is this just an arbitrary list, a grab-bag, or is there some system to it? Is he just trying to come up with some rules that give the answer he wants?
~ What do you make of the rule of attention? That's what really gives his contextualism. Is it well motivated, or just pulled out of a hat?

In response to question 1, I feel Lewis' set rules not arbitrary. I feel as if he is more explaining in detail how people think and not necessarily how we ought to, even though he does produce a theory of how we ought to. I feel his comment on how epistemology and skepticism is too strict for any agent to fully obtain knowledge leaves agents open to fallacious reasoning. Beyond that his suggestions of ignoring and contextualizing your perceptions also leaves the agents open to fallacious reasoning. So for me he removes the hard rock of fallibilism and replaces it with an abyss, pitting the agent in between too strict skepticism and too weak contextualism.

Thoughts?

I agree with this. Although Lewis' rules are an interesting thing to think about, they do not accomplish Lewis' goal of "just barely" dodging both skepticism and fallibilism. As mentioned above they are an insight into how we manage to claim we have knowledge with so many strange possibilities that we are wrong (like the Gettier cases or the evil demon). However, we are still wrong about what we claim to be knowledge all the time. If my professor tells me that he drives the red car in the parking lot I would immediately run home and tell all my friends that I finally found out which car was his. Later I would learn that my professor lied because he was embarrassed of his gross, old brown car and wanted to impress me. We already know that we are right about our knowledge most of the time, the problem is we still don't really know at what point can we claim we have knowledge. Lewis' rules still contain the possibility that we will ignore something important accidentally and thus make our "knowledge" false.

In relation to the second question, I have tried finding Lewis's solution to the actual possibilities (that is, possibilities that actually obtain) that the subject just happens not to know, and therefore cannot turn his or her attention to it. Could one still properly ignore it? That is, the subject may be informed on all other relevant alternatives except for this one actual possibility; but the subject never had a chance to ever even conceive of this possibility, and therefore it has never been brought to his or her attention. Would the two rules then be in conflict? It seems possible (pun not intended) for this situation to occur without breaking any of the other rules that Lewis proposes. Or am I missing something here?

As a parallel of epistemology thinking between Ludwig Wittgenstein, and David Lewis, I would like to refer to the April 16 2010 paper presented by Post Doctoral Researcher Giacomo Sillari, of the University of Pennsylvania. The event was the Synthese Conference, at Columbia University. The title of the conference was Epistomology and Economics.

The title of Dr. Sillari’s paper is: Rule-following as coordination: A game-theoretic approach A few excerpts of Dr Sillari’s paper are as follows:

Make the following experiment: say “It‟s cold here” and mean “It‟s warm here”. Can you do it? Ludwig Wittgenstein, Philosophical Investigations, §510. I can‟t say “it‟s cold here” and mean “it‟s warm here”—at least, not without a little help from my friends. David Lewis, Convention.

"In fact, a different way to state the claim of this article is to say that Wittgensteinian rule-following deals with situations identifiable insofar as a there is a custom. Thus, while not all rules are interpretable as Lewis-conventions, all rules pertinent to Wittgensteinian rule-following16 involve a conventional element and hence can be analyzed as pertaining to situations in which individual preferences regarding their actions are conditional. Such situations are consistent with Lewis‟s analysis of convention in terms of coordination and in fact, as the rest of the article will show, are best understood as recurrent coordination problems.

"Game theory sheds new light on the notoriously obscure pages of the Investigations dealing with rule-following. Taking at face value Wittgenstein‟s indication that following a rule requires that a convention be in place, I have used David Lewis‟s game-theoretic account of convention to clarify how rule-following presupposes agreement an coordination in a community. In so doing, the role played by the community is made more perspicuous, and in particular we have seen that the strategic component is crucial 35 of a full understanding of rule-following. Game theory and the Lewisian analysis of social conventions shed light also on two notions related to rule-following. The notion of Lebensform is illuminated if looked at next to the technical notion of common knowledge, and the notion blind action is clarified in the evolutionary approach. As I have already stated above, I am not claiming that game theory can cover all subtle nuances in Wittgenstein‟s notion of language-game, and neither I claim that hard interpretative issues (for instance that of solipsistic vs. communitarian reading of rule-following) can be settled by game theory once and for all. However, I do believe that I have singled out a group of notions in the Investigations which find precise counterparts in normal game-theoretic ones. Finally, if my analysis does not of course purport to be historical in character, still it highlights that the later Wittgenstein already contains seeds of a philosophy of social sciences that has found voice first in David Lewis‟s seminal study and that, today, continues to grow at the intersection of philosophy and game theory. 36"

I like Dr. Sillari’s claim that David Lewis’s philosophy is an extension of Ludwig Wittgenstein’s philosophy.

The Rule of Belief captures propositions that the speaker never thought of and yet are possibilities(which include actualities according to Lewis). It can be argued that evidence and arguments would back up any possibility P of that nature, whether or not speaker S believes it, but only if we can provide evidence and arguments for every fact of the world, ie. some principle of sufficient reason. So any P which can be supported by evidence and arguments may not be properly ignored. But this would be of help only if not believing P includes ignoring P altogether.

I think the Rule of Attention might be interpreted to be claiming more than it actually is. Lewis is not claiming that if a possibility is not a feature of the conversation context, then it is properly ignored. He is saying only that a possibility may not be properly ignored if it is a feature of the conversational context. And so a possibility P, if supported by arguments and evidence, may not be properly ignored, even if no attention is paid to it.

But you say there may be a fact that, in principle, S could not draw his attention to, meaning it cannot be the case that S had not drawn his attention to P(else that wold be circular). This would happen when certain evidence and arguments are unavailable for S. Maybe it's true in some sense that certain things are beyond S's understanding, therefore he can't grasp them and they would not count as proper justification. I don't buy it though,.

I am troubled by Lewis' web of rules as well. I think the troubles comes from his method in trying to achieve his goals of reconciliation. His methods in many ways want to mirror empirical ways people come to acquire knowledge. He largely shuns armchair philosophizing in favor or armchair psychology. This is troubling in two ways. First, he is assuming that people fundamentally have a good way of acquiring beliefs and knowledge, and that he is simple trying to come up with a coherent way of explaining an existing phenomena in much the way same science explains nature. So he is not proposing some grand theory which if adopted by everyone, would magically the quality of knowledge in human society. Second, he is trying to cover too many moving and contradictory parts at the same time, and these forces are pulling his argument apart. This makes his argument arbitrary. He has all these rules which, while superficially rendering these disparate views compatible, do not do much to explains their overall coherence. The lack of overarching thematic narratives makes the whole project lack rhyme and reason. The faux-empiricism and arbitrariness makes Lewis's paper a confounded enigma.

This is not really in response to either of your questions, but I am uncomfortable with the the rule of conservatism ( and it has nothing to do with my political views). Lewis suggests that the rule of conservatism can be derived from the rule of reliability and vice versa, I am not sure I agree. I agree that the rule of reliability could be derived from the rule of conservatism, however I am not sure it works the other way around. This is likely because of my discomfort with the conservatism rule. While I am willing to ignore many possibilities, in certain contexts, because doing so allows communication to occur, such as the examples regarding the use of words like: all, never, every and none. However depending on the context one could be required to consider truly absurd possibilities, or ignore reality, the latter is dealt with by the rule of actuality. The former though that one could be required to consider the ridiculous bothers me and is also a problem with the rule of attentiveness. As a result of either rule, if I were to attend a starwars or bigfoot convention I would be forced to consider the possibilities that bigfoot exists and starwars may have actually happened a long time ago and in a galaxy far far away. And while I think Lewis may be able to do without the rule of conservatism he seems commited to the rule of attention, for that reason contextualism is unappealing to my sceptics heart.

I also feel unsure about Lewis' rule of conservatism. From what I understood it sounded like this rule allows one to properly ignore certain possibilities if others commonly ignore it as well. It might be a misunderstanding of this rule on my part, but this seems to weaken and if not undermine some of Lewis' other rules. If one allows common knowledge to guide their reasoning then it seems that it would just leave one with assumptions, rather than any true knowledge. Lewis mentioned how the rule of conservatism can be derived from other rules and vice versa, but I was unclear of whether this rule could stand on it's own when properly ignoring a possibility.

In response to question 2, "What do you make of the rule of attention? That's what really gives his contextualism. Is it well motivated, or just pulled out of a hat?"

Lewis' Rule of Attention maintains that any possibility that is ignored is properly ignored. He then says that a possibility not ignored (or relevant alternative) is not properly ignored. The problem I find with this is that any possibility can potentially be brought to attention in some conversational context. So it seems to me that any possibility that Lewis would call 'rightfully ignored' is really just a possibility that hasn't yet been brought to attention. But given enough time all possibilities could eventually be brought to attention and would therefore be considered relevant alternatives according to Lewis' own definition. Since any possibility brought to attention becomes a relevant alternative what could be left as properly ignored or an 'irrelevant alternative'?

The way he resolves this is by saying if an unwanted possibility comes into conversation "we might quickly strike a tacit agreement to speak just as if we were ignoring it and after just a little of this, doubtless it would really be ignored." Maybe Lewis and I have different understandings of the word 'ignore', but if one is 'actively ignoring' a possibility surely he has paid attention to it at one time instantly transforming it into a relevant alternative, a process that I doubt can be undone based on Lewis' prior definition of a possibility 'not properly ignored' as a possibility 'not ignored'.

I'm fine with the world being one big guess, a guess contextualized alongside the guesses of yourself and others, with I guess, there also being a wonderfully complex orgainicly fallibil system that those guesses are bound to. That does not mean that all claims are equal, it just mean that none are foundational. Our best guesses are the ones we simply can't ignore becuase their 'truth' permeates so throughly our other guesses, they support them and may be the reason we conclude to 'know' other more periferial guess to be true. I'd say that there are three things that are definately true, but they unfortunately get you no were on thier own, so though solid, they are no foundation. They are (1) there is existence, (2) there is thought and (3) there is perception; two and three might be the same thing, and if you can build anything as definite as those three out of those three, please do. Otherwise, may the best guess win.

He (Lewis) definitely is just trying to come up with some rules that give the answer he wants. Those rules, because they are incredibly difficult for me to make sense of, are arbitrary.

• What do you make of the rule of attention? That's what really gives his contextualism. Is it well motivated, or just pulled out of a hat?

I think what Lewis says about the rule of attention is well motivated and not just pulled out of a hat. He says at the end of p. 599 "To investigate ignoring of [possibilities is in fact] not to ignore them...Knowledge is elusive. Examine it, and straight away it vanishes". This is very well said and we see this whenever we try to define what knowledge truly is. Is it justified true belief? well no that was clearly proven insufficient by the Gettier problems. Is there one word or a sentence tat can describe what knowledge is? Is there a definition? It seems practically impossible to define knowledge and the closer we look at it the more certain we become that we have no knowledge in the first place. In the beginning he started to talk about all the things that we do know. And it seemed as though humans in general have a vast array of knowledge of all kinds of things but as we tried to examine these things more closely we became uncertain of them all and it seems that all we can truly know as Descartes says is that we ourselves exist. Nothing else is certain and can be proven without a doubt. So in order to surpass this problem of skepticism of now knowing anything we can apply Lewis rule of attention and choose to ignore possibilities which are that are most likely not true (the evil demons deceiving us). And by doing so we can at least come up with some sort of explanation of what knowledge is.

Lewis's rules do not seem to capture a very solid reasoning behind contextualism. Yes, his modality and Rule of Actuality seem plausible, with uneliminated possibilities highlighting the importance of the spatial and temporal features of experience. The Rule of Belief neglects to state a sufficiently high possibilities to be properly ignored, beyond actuality. The Rule of Attention I find difficult to understand, attending to context-dependence of making an additional comment to another that was previously said. This seems to break any sense of closure, based on its context in the conversation. Where does is the line drawn between what are purely thoughts and what is said in the Rule of Attention? At this point is it properly ignoring a possibility if it is not said? Bound by these conversational suggestions, it appears to not account for mistakes, impulsivity or useless statements in such that they are of equal relevance now to the conversation. (But, perhaps, this is exactly what Lewis was trying to argue with his nearest possible worlds, in that they are in fact a possibility.)