Of Manifold Discourse
Spies and Such
The examples goes as follows. It is supposed to show how problematic it is for us to quantify into propositional attitude contexts.
Smith lives in an apartment building. He sees a shady figure down the hall. Smith, a paranoid sort, comes to believe that (1) the shady figure is a spy. Later that day, Smith attends a dinner party. Smith meets every person at the party, in particular, the host. Smith finds the host to be a good natured chap. He comes to the belief that (2) this man is not a spy. To put it another way, Smith does not believe this person is a spy. However, as we are philosophers, supposing to our heart’s content, let us further suppose that the shady figure and the host of the party are in fact the same person. Thus, it turns out this person, X, bears to properties: the properties of (i) Smith believes X is a spy and (ii) Smith does not believe X is a spy. “Smith does not believe X is a spy” is claimed to follow from “Smith believes X is not a spy.” Thus, we have a contradiction in properties that we predicate of our subject, that is, X.
Quine’s response is: Do not quantify into propositional attitude contexts (or opaque contexts, environments).
One response I have encountered is that it is fine for Smith to have contradictory beliefs. We “adjust” our beliefs or assent to correction all the time. This seems fine, but does not directly confute the issue of quantification over propositional attitudes being problematic. Granted, Smith is missing an important fact: he does not know that both spy and host are the same person. However, we are not dealing with knowledge. For one, if we accept justified true belief as a conception of knowledge (ignoring a fourth condition and Gettier-thinking), then knowing this fact must presuppose that one at first believes it, the fact being that they are the same person. If he knows it, how did he come to know it? Well, clearly if you introduced him to the fact, he likely would not assent to it (supposing that the host seems to him convincingly non-spy-like). So he might not believe him. But in order for him to know it, he must at first believe it. Will the evidence (justification) convince him if he does not believe it? It being true seems to hold no weight either if Smith is steadfast in his belief. So revealing this fact seems to do nothing for our situation. I think the aim of the JTB route (or any gripes about Smith later coming to just know rather than believes) carries an implicit premise that having contradictory beliefs is acceptable, given that we revise them when we are given new evidence, and that the contradiction in quantification over propositional contexts, if only in this example at least, evaporates at the same time that Smith realizes X1 (spy-guy) and X2 (most host) refer to X. In a strong sense, this conception holds that status of contradiction within the proposition tracks the status of the contradiction outside the belief-context (at the level of quantifying into it). Others do not think this is the case.
(a) “… is a spy” and “… is not a spy” are not contradictory. (b) “… believes that …” and “… does not believe that …” is not our major concern; though, suppose on the phone Smith recounts the events of his day. He says to his friend, “I saw a spy … I met an excellent host who I would not believe is a spy.” The proponent of this quantification problem, to show that it is a true problem, will say the epistemic possibility of Smith believing that X changes the problem, does not satisfy us if we accept JTB, or might simply be rejected by Smith. We’re left at the contradictory properties to which collect under X. This leaves a problem for us because we cannot consistently talk about Smith’s beliefs (if we generalize to various other similar situations).
With luck, this sums up the problem. I’ll saw my bit on it. Perhaps to your surprise, my remarks will seem like arguments in favor of the “tracking” retort. I’m still working things out. My idea is this: Smith perceives X1. The only predicates Smith has for X1 are “is a spy” or “is a shady figure.” For the later encounter with X2, Smith has likely “is a good host” and negatively “is not a spy.” Again, presumably, the predicate “is not a spy” directly translates into a negative belief: “… does not believe X is Y.” I’m really sketchy on how this might look in the predicate calculus, so I cannot suffer a logical showing of this. But we can suppose it possible and valid for our purposes.
My argument has to deal with the representational nature of the contradictory belief statements attributed to Smith. It seems that at the level of belief (Smith believes that…), there is an object within Smith’s proposition. (Perhaps my analysis is Russellian.) Smith is asserting, “There is some X” in believing that X1 is a spy. In this case, Smith seems to never be talking about or referring to the same object when he utters his second belief about X2. His two individual beliefs assert the existence of, and are constrained by, the totality of predicates Smith would assent to of each person (X1 and X2). At the level of proposition, separate objects are therein. At the level of proposition, and due to Smith’s evidential contraint (the certain properties manifest to him), Smith must have two disparate entities (X1 and X2) for his propositions. His mental states, in this case, are about two encounters and thus two individuals. It seems that this cannot be got around at the level of proposition.
In this case, it seems that if the propositions have differing content, they cannot be in contradiction. The predicates (as far as Smith would assert to) constrain the identity of the disparate subjects of those propositions. So, “… is a spy” constrains Smith to believe something like “Spy-guy is a spy” or less analytically “Spy-guy is a shady fellow.” And separately, “Bob, the host, is a good host.” For Smith, both propositions have content which contain different objects, so restricted by the predicates Smith would assent to of those objects.
London versus Londres (and modified)
Suppose Pierre is a monolingual French speaker who lives in a rural town near the French Alps. One day he picks up a magazine titled Londres Aujoud’hui!, and he flips through the magazine. He sees photography of London, England, and comes to the resolute belief that London is pretty. Since he’s a monolingual French speaker, he expresses this belief in the French as “Londres est jolie.” Pierre returns to his house. A few months pass, and we now see Pierre packing for a journey: he’s going to tour the world. He’s got a pipe dream conviction to do so. He stows away on a run-down small ship, packed in a heap of salmon. In a few days, a few miles of bus riding, he winds up in London. Quickly, he settles down and starts to learn the language of the Londoners: some variant of English (something we’d likely interpret more or less accurately).
Now, after about 3 months of being Londonized, Pierre calls his friend back at home: “Non, je n’aime pas cet locale. Mais, je dois trouver la London! Londres est jolie!” Unbeknownst to Pierre, however, he is in fact in Londres. He just does not know “Londres” and “London” are the names for the same city. So, does he believe that London is beautiful? (Kripke asks with a certain stubbornness.)
Most would argue that, “hey, once he gets to London, that overrules his previous French expression. It’s just an expression. It can be overruled.” However, the same situation maintains of “facts” that was present in the last situation. Pierre learns no new fact to which his old belief might be revised, and positing that he should just be told the fact, or that it might be handed to him by God, perhaps, changes the situation we’re talking about: on the phone, he’s in a state of mind that has contradictory beliefs, it seems, and we cannot for the life of us quantify over it. I try to think of it think way, though it may mess up the example. Pierre has a French-Mental-State (FMS) and an English-Mental-State (EMS). This is what it seems like Kripke is trying to pin us down to, given Pierre’s performance. Pierre is “thinking in French” in one case and the referent of his French use seems to be something other than the referent of his English (or Londoner) use, where he’s “thinking in English.”
One issue many suppose is that “well, now he’s in London. Him being in London in fact introduces him to the real London whereas back home he was not properly acquainted.” I doubt this argument really does much for we can assume in my modified version that Pierre took the trip first then picked up a mag at a French shoppe before becoming fluent (to whatever extent) in English. In this case, he’ll utter “Londres est jolie!” and no one will know what he’s talking about (and they won’t be able to correct and give the co-referential fact to him because, perhaps, they don’t speak a lick of French).
Another issue is the predicate “is pretty.” One initial complaint in my philosophy of language class was that the predicate “is pretty” has no extension. There is no set of pretty things. This may sound depressing but it seems intuitive when one considers a view sub specie aeternitatis. However, surely, we might say “pretty relative to culture X.” Perhaps. But we can use a different predicate such as “is between X and Y.” Perhaps not… I’m losing an example that would actually count as a genuine belief. It seems that when we get into extensional (in principle empirical claims) we reach a problem of the Sorite’s paradox (which I will discuss later). It has a strong and a weak sense. The weak, which I’ll discuss now, has to do with saying of Pierre’s belief that its Sun rises at a certain time. Clearly if this is the case, it would be had to imagine, for our sake, Pierre maintaining such a trivial belief. But further, he would easily be able to falsify it and perhaps use inductive reasoning to conclude that he is in London after all. It seems that we’re highly depending on the vague nature of “pretty” in that it allows us to step beyond any naturalistic definition. (I am thinking the naturalistic fallacy is relevant here. Can we say the “is pretty” follows, as a normative or aesthetic term, from non-normative or non-aesthetic things necessarily.) We might be able to say for Pierre, but we’d certainly have no definition of our own to work from. However, that might just be all we need: Pierre’s intensional definition of “pretty,” what it means for something, in qualitative terms, to be pretty to Pierre. But it seems we’re back our own “Sun rising” problem. This leads me to the strong Sorites problem.
First I want to get to my initial argument, though, about the words “London” and “Londres.” At first I was thinking in terms of Kripke as far as Pierre appropriating these names in a causally appropriate way. At first, I just though he got two names from disparate initial baptisms. I went further on to think that these linguistic entities, though they may be lexically different and phonetically different, they are not in language the way “est jolie” and “is pretty” are. First off, there’s a translation rule which says how to properly grammatically connect the copula “is” and “est” with the property “jolie” and “pretty” to form grammatical predicates. In the most superficial sense, the letter arrangements represent English and French. In this sense, “London” and “Londres” are in language, but I do not think names, in general, have places within the grammar of language. Nouns have no special status in that they can be replaced in any predicative statement and while the statement might change in meaning, it will still be grammatical. But what else can you do with a noun? Well, you can use nouns relationally with prepositions, but you can do this with any noun. This just seems to be a feature of nouns. I am interested in a specific feature to names that seems to not be restricted by Kripke’s notion of rigid designators, which is what “London” and “Londres” seem to be (at worlds in which the city exists). We might refer to London with a pronoun, eliminating in certain cases its proper name. What I am getting at is that “London” does not seem to be in the English language as if Pierre is speaking EMS. “London” and “Londres” are not contrasting names of the same object. This conclusion seems wildly in contradiction to the scenario Pierre plays out on the phone with his friend, for it seems that he is using two different names for the same object. (I’m really not sure just how ineffectual this idea is. But it piques me nevertheless.) It seems like these two names just mean “that place over there!”, but in Pierre’s case it is an empirical matter… He simply did not know where London was or where he was going.
But we do not want the dualism, to say he “believes in the existence of two Londons.” There cannot be a London and a Londres to which correspond to Pierre’s separate expressions. But thinking this seems to presuppose that Pierre was referring to the same place in both instances of uttering his beliefs. In the latter case, he does not know he’s in that place to which he is referring. I’m thinking here that when Pierre says “Londres est jolie” for the first time he’s considering himself as though he were there in London after walking through a special door titled “Londres” in his mind. This may also help eliminate the idea of his not being acquainted.
I suppose I will get to my point now… It’s related to the Sorites paradox. I have a problem with us being forced to believe that “London” and “Londres” refer to a location with rigid borders. I am thinking that Pierre may have a mobile phone. He may walk to the outskirts of London and say “Cette locale est moche.” Will he be talking about London still? Can we verify this? How much of the outskirts of London is actually London? If we suppose a rigid geographic border (and draw a line that demarcates London from say another locale), we could put Pierre on that border. If he steps over it, will he still be in London? Does London expand with his steps and the belief (mental state) that affirms that he is still in London? Is London’s border just a product of my mental state? Will he be in contradiction to his previously asserted belief if he actually steps outside of London by an inch?
Further, to which section of London did the pictures he initially saw capture? It seems like a bold presupposition to make that the words “London” and “Londres” must refer to all parts of town for every speaker. So we suppose that Pierre read about Downtown London but decided to proclaim something about all of London. Well, this just makes no sense. Surely he would not assent to the claim, “So you believe all of London is pretty based on what you just read about it?” Might he admit the possibility of error? Is it fair to say “suppose the error away”? I do not think you can “hold him to his word.” Again, if he’s in a town, would he say the whole town is “ugly” if he very well knew he had not seen it all. My point is that given these facts, we seem to be in a bind with even attending to our variables for quantification.
It seems that “London” refers to a geographic approximation. What if the place he lives in is called “London” by all the speakers for short of “Lesser London” because the inhabitants do not want to always invoke bad feelings? They want to be closer to the image of London. And suppose the photos of London were in fact photos of London still. Suppose further than Lesser London is very geographically close to London and it only recently was dismembered because the Londonites wanted to be rid of the trash. This occurs during Pierre’s trek. I suppose this changes the situation too much.
I do not think these names can refer to a name as though it were not already defined. But did Pierre define the borders of London?
It seems like the two belief statements are being treated as being ontologically different when they are in fact only epistemologically different. Take the representationalist view again: there’s an object of London in the proposition with a certain set of features from the photo… This seems like the “acquaintance” again…