nf0098: er where are you going sf0099: Cornwall nf0098: oh that can't be any better than the Philosophy weekend sf0099: well no but it's paid for [laugh] i might come for the day on Friday nf0098: yeah good plan okay then anyway perhaps we should get started so we're looking at proper names at the moment right remember from last week and we started off last week by looking at a Millian theory of proper names anyone want to remind me what a Millian theory of proper names is without looking at your handout 'cause that would make me really depressed if you can't remember what a Millian theory of names is without looking at your handout namex sm0100: that they are are labels nf0098: they're labels exactly right so what's the meaning of a proper name for Mill sm0101: the er hook they cos-, they correspond to in the er world nf0098: yeah exactly the object that it hooks up with in the world so for Mill the answer to what is the meaning of a proper name is very very simple it's just the object in the world that it hooks up with and we finished off last week by looking at three worries with that kind of Millian theory of names anyone want to remind me what the problems were sm0100: i don't know fake names like nf0098: exactly right not tend to call them empty names sm0100: empty names nf0098: names that don't have anything relating to them in the world exactly right namex problem with empty names er there was a problem with how we decide on the denotation remember the Millian difference between denotation and connotation denotation being the object in the world connotation being a kind of a property er and if we think that names are things which just have denotation and don't have connotation we've got a trouble on settling or determining the referent how do we find out what thing it is in the world that the name is supposed to hook up with anyway and one more problem anyone remind me of a last problem sm0102: the Hesperus and Phosphorus problem nf0098: Hesperus and Phosphorus exactly right there's a problem with what i called cognitive difference okay and that was the worry about think if all there is to the meaning of a name is just the thing in the world that it hooks up with how do we explain the fact that sometimes [sniff] er you can not realize that fact sometimes you can think things like Hesperus is not Phosphorus even though in fact those two names pick out exactly the same object in the world okay and it's this problem the problem of so-called cognitive difference that we're going to concentrate on for the next sort of week and a half [sniff] and we're going to concentrate on it because this kind of problem was first brought to the fore and made most of by Frege so we're going to see how Frege established this problem and what sort of theory it led him to develop for names it's going to be a non-Millian theory of names that Frege developed in response to this kind of worry okay then so what exactly is the problem of cognitive difference then what is the worry here well i've said on the handout some sentences which state sameness of referent like Cicero is Cicero or the first postmaster general was the first postmaster general seem very uninformative [sniff] you know if i stand up here and lecture to you and tell you hey Cicero was Cicero you're going to think you're being sold short you know that's not particularly informative or useful thing er for a university education to equip you with but on the other hand if i stand up here and tell you Cicero was Tully or the first postmaster general was the inventor of bifocals or even Abraham Lincoln was the first postmaster general or the first er the inventor of bifocals you've learned something useful right fairly useful anyway seems like taxpayers' money is not going to waste here you've learned something got a useful fact to take away with you but how can the Millian explain something like that because according to the Millian the meaning of the name is entirely given by its referent hello namex recovered from your hangover leaving the union at two o'clock in the morning [laughter] sm0103: yes nf0098: yes [laughter] very good and it's only taken you till two o'clock the next day that's fantastic [laughter] er so if the Millian is going to tell you that the meaning of the name oh can't believe it i got a new pen the other day and it's run out already er the meaning of the name Cicero is just given by your little Greek orator there he is meaning of the name Cicero given by your Greek orator how does he explain the fact that it can be useful then to learn that Cicero is Tully because if the meaning of those two names is entirely given by the object they pick out in the world then you know how is one any more informative than the other how would Cicero is Tully be any more informative than Cicero is Cicero okay so i say on the handout this difference would seem inexplicable if all there was to the meaning of a name is the object it refers to and i say this is sometimes known as Frege's problem or the problem of cognitive difference well we might perhaps note here that i'm being just slightly disingenuous i guess because i'm saying the problem as if there's a single kind of problem we can identify here when actually it's more like a kind of bundle of worries because if you read any of the literature here you'll find that this problem is sometimes stated in slightly different ways okay so i've said that it's you know what we're going to call Frege's problem is the difference between identity statements of the form A-equals-A and identity statements of the form A-equals-B and that's one way of stating the problem but in other areas you might find the problem stated as er see er the fact that it might be rational for someone to believe okay say that the object named by the label A is the object named by the label B would it be rational could it be rational to believe that A is not B yes even after nineteen pints namex can see that [laughter] it would be rational to believe that kind of thing would it be rational to believe that A is not A do you think no very good wouldn't be rational doesn't seem to matter how confused you are about your environment it never seems to be rational to believe something of this form that an identity sign using the same label on both sides for the same object er that those two objects are non-identical so sometimes Frege's problem is stated as the fact that it would be irrational to believe something in form A is A A is not A but irrational to believe something in form A is not B sorry did i give that the wrong way round [laughter] sf0104: we understood nf0098: other way round you know what i'm saying er and a third way that you might find this problem stated is to do with exchange in certain contexts okay so in certain contexts it looks as if you can always exchange coreferential terms what's called salva veritate get all classical on you there did we do this last term did we do the term salva veritate sm0105: er sf0106: er it was mentioned nf0098: that was a quick statement for no [laughter] no certainly didn't do that what the hell is it [laugh] salve veritate means exchange er without alteration in truth value if you can exchange something salva veritate that means that you can exchange two things without altering the truth value of the sentence in which they occur so do you remember all the stuff you did in er logic in the second year you did stuff about truth functional connectives like and and or and not things of that form you can always exchange coreferring terms salva veritate if you've got a sentence like Cicero er went to the market [sniff] or i don't know er Cicero went to the market or Caesar did you can always exchange coreferential terms in this kind of context without altering the truth value if this whole sentence is true then the sentence Tully went to the market or Caesar did has to be true as well if Cicero is Tully yeah 'cause these guys like or that's truth functional all they worry about is the truth value of the sentences they're connecting and if it's true that Cicero went to the market then it has to be true that Tully went to the market if Cicero is Tully so in some contexts then you're guaranteed to be able to exchange coreferring terms salva veritate but in some contexts you're not anyone think of a context in which you might not be able to do that sf0107: John believes that Cicero nf0098: yeah classic example if you've got something like John believes that Cicero went to the market [sniff] something like this a so-called intentional context using a term like believes or hopes or desires a term picking out a mental state er a propositional attitude in contexts like these you pretty obviously can't go around exchanging coreferential terms salva veritate John believes that Cicero went to the market so this whole sentence is true okay this whole thing is true am i guaranteed that the sentence John believes that Tully went to the market is also true sm0108: no nf0098: no exactly right not guaranteed that at all John could easily believe that Cicero went to the market and disbelieve that Tully went to the market precisely if he doesn't know that Cicero is Tully so in contexts like these believes contexts hopes contexts desires contexts things like that it doesn't look as if we can exchange coreferring terms salva veritate we can change the truth value of the whole sentence by swapping two terms which pick out the same object so this is another way that you'll find Frege's point made Frege's argument made because obviously there would be nothing to explain this difference in the kind of context where you can exchange coreferring terms if all there is to the meaning of a proper name is its referent the meaning of Cicero is completely exhausted by our Roman orator how are we going to explain the fact that you can expla-, you can exchange the terms Cicero and Tully in this context but you can't exchange them in this context it looks as if you're going to have to introduce some other element into the meaning of the name to explain this variation okay do people see that namex mm vaguely namex you're looking kind of dazed [laughter] is that just because it's sm0100: that's just me usually nf0098: week three of term just the usual okay okay cool right so those are three different ways that we might find Frege's problem stated there's the idea of informative identity statements that it's better to learn something of the form A-equals-B than it is to learn something of the form A-equals-A er there's the fact that it's sometimes irrational to believe something of the form A-does- not-equal-A even though it's rational to believe that A-does-not-equal-B and we've got the fact that in some contexts you can exchange two coreferential terms you can exchange A and B but in other contexts it looks like you can't so this this whole kind of cluster of worries that often get referred to as Frege's problem or as the problem of cognitive difference so that's a whole kind of cluster of worries but i put it to you if we're going to be driven from our nice simple intuitive clear er Millian position on names we want more than just a kind of cluster of you know general slight worries about what's going on here we want an argument goddamn it we're philosophers we want a nice clear argument that shows us why we can't be Millians given all these states of affairs and actually i think it's harder to get an argument in this case than we might initially expect devil's in the detail like it always is you know so how might we formalize the kind of argument that's going on in cases like the ones we just looked at well i've given you a kind of example on the handout or a way we might try it on the handout we start off with a piece of data right the data is say that senten-, the sentence Hesperus is cold and the sentence Phosphorus in cold is cold differ in cognitive value because a rational agent can know both and assent to one while dissenting to the other so there's a piece of data for you and that seems er incontrovertible right that's just clearly right now secondly if that's our premise so if that's our piece of data we need some kind of second premise connecting the objects of knowledge and the expressions of our language because look what premise one tells us is that an agent can know both of these things and can assent to one and dissent to the other the first premise is one about epistemology right it's one about what people know but what we're trying to get at is a conclusion about the meaning of linguistic items so we need a premise connecting the things that people know about bits of language to the meaning of bits of language and we could just do that in some very general way by saying there is some kind of connection between the objects of knowledge and expressions in the language now if we can establish that if we can establish that there is this connection then we do seem able to draw a conclusion that Hesperus the term Hesperus and the term Phosphorus differ in something other than and i've put in here bedeutung that's the German phrase that Frege uses sometimes translated as reference sometimes as nominatum sometimes as designation i'll almost always say reference but you should just be aware that that's er an issue about exactly how that term should be translated okay we can obviously get that conclusion because if you've got the two sentences Hesperus is cold and Phosphorus is cold the only bit they differ in is the names they contain yeah so we can trace if there is going to be a difference er between the meaning of those two whole sentences we know it's got to come in the meaning of those two bits of sentences so we do get the conclusion that Hesperus and Phosphorus differ in something other than reference that would seem to be a good conclusion 'cause that's what Frege wants to conclude because he's going to say that the meaning of name isn't just its reference but i think you should know that as it stands the conclusion in three isn't particularly interesting oops [laughter] you can't get that back now either 'cause that's stuck in the middle [laughter] going to have to go under the table Andrew very good okay so the conclusion in three isn't an interesting conclusion why not why isn't it interesting to be told that Hesperus and Phosphorus differ in something other than reference well the answer is clearly why it's not interesting to be told that Hesperus and Phosphorus differ in something other than reference it's 'cause that they clearly differ in something they're not the same thing are they this isn't like saying Hesperus and Hesperus differ in something other than reference these are two different syntactic items they've got different orthographic properties they've got different phonetic properties they've got different properties about occasions of use there are lots of different properties that these have regardless of the fact that they don't differ in reference so what Frege needs to make this conclusion of his argument an interesting conclusion is not the claim just that they differ in something other than reference but that they differ in meaning that's the interesting conclusion that Hesperus and Phosphorus differ in meaning because if that's the case then we know that meaning can't just be reference it can't be the Millian position because we know that Hesperus and Phosphorus both pick out Venus yet again this week looks a bit like Mercury but anyway we know they both have the same reference if we know that they both have the same reference er but yet they differ in meaning then we know that meaning and reference aren't synonymous aren't the same thing so that's the conclusion that Frege needs to get whoops he needs to get that Hesperus and Phosphorus differ in meaning and that's a claim that's been disputed by some people some philosophers want to accept the initial kind of uninteresting claim that i put up to begin with that Hesper and Hesperus and Phosphorus differ in something other than reference and yet not accept that they differ in meaning I- E not accept that there is a semantic level difference between these two guys how might you do that well you might do that by appealing say to the syntactic properties of each expression or i guess a probably more popular move is to appeal to the kind of pragmatic properties that these expressions have the fact that on some occasions of use it's fitting to use the term Hesperus and on some occasions of use it's fitting to use the word Phosphorus so what these kind of people like er Nathan Salmon wrote a whole book Nathan Salmon one of the great fish philosophers of the century did we look at him last term no er Nathan Salmon wrote a whole book called Frege's Puzzle in which he argues that the only difference that's relevant here between Hesperus and Phosphorus is a pragmatic difference why don't you say something like the ancient astronomers believed Hesperus was Phosphorus okay it was the ancient astronomers that introduced the two different terms Hesperus and Phosphorus one for the star that you could see in the morning one for the star that you could see in the evening so they introduced these two terms they didn't realize that they were naming the same planet 'cause obviously that wouldn't have been very helpful to have used two different words for something they realized was the same planet they introduced two different words for it called one Hesperus and one Phosphorus er so it doesn't sound right to say that the ancient astronomers believed that Hesperus was Hesperus even though it does seem right to say that the ancient astronomers believed that Hesperus was Hesperus that's the kind of failure of exchange that w-, Frege's Puzzle is interested in but people like Nathan Salmon want to say well look actually you can say both those things both of those sentences are perfectly r-, right they're perfectly true because semantics is just to do with reference it's just that this one this first sentence is not very acceptable there are pragmatic reasons that you don't want to er attribute this kind of thing primarily because this isn't something that the ancient astronomers themselves would have assented to even though when they were believing that Hesperus was Hesperus they were really believing that Hesperus was Phosphorus that's not the way they would have chosen to phrase their beliefs now i can see certain people like namex [laughter] and namex not looking totally convinced by this manoeuvre here and i i'm in sympathy with you really it doesn't sound like an intuitively very compelling manoeuvre although you know Salmon's book's worth having a look at because he gives this kind of line a really good go you know it's a big book lots of arguments for his position but if you're feeling a bit kind of sceptical about this sort of worry then i think that's fair enough really and maybe if that's the way you are feeling then we can kind of let Frege off the hook even if we can't find a really strong way to kind of codify his argument perhaps it is the case that the worries he raises are good enough that if he can get to the conclusion that Hesperus and Phosphorus differ in something other than referent we just kind of feel that has to be something to do with the meaning of the expressions involved even if he can't actually get you to that conclusion deductively perhaps it just seems reasonable that the thing that they differ in has to be a thing to do with the meaning it can't be something to do just with the pragmatics as Salmon tries to argue okay how are we doing people clear on the kind of argument yeah sf0109: can i just ask is the nf0098: yeah sf0109: is the Salmon position something we really need to sort of know about in detail nf0098: no no sf0109: if you er er nf0098: it's if you're interested you should have a look at it sf0109: mm nf0098: but it's not sf0109: it's not going to be an exam question or anything [laughter] nf0098: oh God what sort of question is that i'm not answering questions like that [laughter] okay then so that's Frege's argument this kind of cluster of worries that seem to show us that the Millian position which tells us that the meaning of a proper name is just given by the object it refers to that that position is somehow unsatisfactory so clearly we know that position is unsatisfactory and we've got a good argument against it what we need to do next is to come up with a better position we need to find a positive suggestion as to what the meaning of a proper name is which is going to allow us to avoid these kinds of difficulties and Frege's got a solution and being a great philosopher he's got two solutions one of which he made in an early stage of his career went on to reject and the second one which has become a kind of standard of philosophy of language so his first suggestion which he made in Begriffsschrift one of his earlier works was that a sentence like Hesperus is Phosphorus should be read as saying something like the object denoted by the expression Hesperus is the same as the object denoted by the expression Phosphorus okay you all remember from last term notice the way that's not a question that's a statement you do all remember from last term i know you do the difference between using and mentioning anyone want to remind me what the difference between using an expression and mentioning an expression is we did it in the first week of the first term so you know way back [laugh] namex sm0103: er well one of them's kind of quotation marks round it er which means that you're referring to the word nf0098: mm-hmm sm0103: rather than what the word refers to nf0098: exactly right perfect the difference between saying something like John is happy which tells you [sniff] that right and saying er that's single quotes which tells you what sm0110: the word John is that nf0098: yeah doesn't really make much sense this tells you something about the word because you've enclosed it in quote marks you're mentioning it you're not using it you're talking not about this big eared guy up here with a grin on his face you're talking about the linguistic item the word itself and you can use it to say true things like like that John has four letters that's fine and if you take off the quote mark what does that mean sf0111: the person John has four letters nf0098: exactly maybe that's why he's so happy [laughter] it's his birthday he's got four cards [laugh] er so Frege's first manoeuvre here is try to differentiate use and mention what he's saying really is that when you have a sentence like Hesperus is Phosphorus you're really not saying something about the world you're saying something about your language what you're saying is that the object referred to by that is the object referred to by that okay put your name Hesperus and your name Phosphorus you've enclosed them in quote marks so you're mentioning them you're not using them and you're saying that the thing which is picked out by this linguistic item is the same as the thing picked out by this linguistic item now this was Frege's first proposal but he rejected it pretty quickly oh we might talk about this as a metalinguistic proposal okay because it's explaining a linguistic fact by using items from the language in the explanation [sniff] it's kind of going one level higher than language right it's metalinguistic because it's mentioning bits of the linguistic system so this was his first proposal but he rejected it pretty quickly because he just thought that actually on reflection this whole claim here just isn't what's meant by a claim like Hesperus is Phosphorus he just decided that this equivalence claim to do with bits of your language isn't what's meant by a claim like this without the quote marks and he thought that because if you look at a sentence like Hesperus is Phosphorus or a sentence like er Abraham Lincoln was the first postmaster general or Cicero was Tully those claims don't seem to be telling you something about your language i'm guessing that none of you knew that Abraham Lincoln was the inventor of the bifocals when you came in here sf0112: Benjamin Franklin nf0098: was it was wasn't it sf0112: nf0098: you're absolutely [laughter] right oh my goodness [laughter] sm0113: that's why none of us knew it i think nf0098: [laughter] that's very true or i could have bluffed my whole way through here [laughter] and you could have got a degree from a university and gone out with a completely false piece of knowledge [laughter] couldn't i absolutely right Benjamin Franklin this is the this is the example gets used in all the books as well so it shows how carefully i've been reading things [laughter] Benjamin Franklin indeed was the inventor of the bifocals so namex block your ears now but i'm assuming some people in fact nobody knew that Abraham Lincoln was [laughter] the inventor of bifocals when you came in here so that's a good example [laugh] er but when you learn something of that form it doesn't seem like you learn something about your language it looks like you learn something about the world you learn a genuine useful fact to go out and apply in the world and not just something to do with your language and because of that Frege thought that this kind of restating of the problem in terms of a comment about the language wasn't a good way to go so even though this sort of approach might solve the problem he just thought it was getting things wrong okay so once he'd rejected this sort of proposal what's he going to do well being a very smart guy he's going to come up with a whole new theory of the meaning of proper names according to Frege in his classic paper Sinn und Bedeutung he's going to tell you that there are two elements to the meaning of a proper name Mill tried to tell you that there was just one there was just this guy in the world the denotation or the referent but what Frege's going to tell you is that although this is right although this forms one element of the meaning of the name there's another element and that element is the sense sometimes also called the mode of presentation whoops so for Frege the meaning of a proper name is a kind of bifurcated thing it's got two elements to it a dual er feature two features on the one hand you've got the object referred to in the world the referent or the denotation and on the other hand you've got the way in which that object is presented so think about Hesperus and Phosphorus again you might think that the sense of Hesperus is the bright star seen in the morning and you might think the sense of Phosphorus is the bright star seen in the evening even though the referent is the same for both of those expressions or here you might think that the referent of Cicero is our Greek orator but the sense is not entirely sure what the sense of Cicero would be er the author of certain works and it's going to be this element that's going to explain why it is that we get cognitive difference okay and why is it that you can believe Superman can fly when you don't believe that Clark Kent can fly well that's because even though you've got the same referent in the world for Clark Kent you've got a sense to do with wearing glasses being a bit of a skinny wimp working for a newspaper whereas for Superman you've got a sense to do with wearing a cape wearing his underwear outside his trousers flying around flying that's pretty good you don't get flying with Clark Kent so even though you've got the same object being presented in both these cases you've got radically different kind of modes of presentation of that object namex you're looking sf0112: so are things like definite descriptions like the first postmaster general or something like that are they senses or are they names nf0098: well we're going to need to look more carefully at exactly what a sense is here sf0112: mm nf0098: i'm trying to give you a kind of rough idea to start with and then we're going to try and spell out what the technical notion of a sense is the question about definite descriptions er we need to remember we're talking about proper names er in a kind of non-technical sense at the moment it was quite clear when we looked at Mill that he had a fairly restricted notion of proper names but Frege has a much more liberal notion sf0112: mm nf0098: okay Frege's going to tell you that something like Cicero is a proper name something like Clark Kent is a proper name something like Superman is a proper name he's also going to tell you though that something like the first postmaster general sf0112: mm nf0098: is a proper name because he's got a class of what he calls eigename eigenamen which contains both ordinary proper names and definite descriptions sf0112: does that mean then when we're trying to sort of work out how to express the sense of Cicero we couldn't use something which is like unique to Cicero because if we did nf0098: well sf0112: wouldn't it be a proper name nf0098: er sf0112: as the sense nf0098: there's going to be a question about whether senses are just forms of definite description sf0112: right nf0098: some people read them as just being forms of definite description although even then you might want to be careful because er talking about replacing a name like Cicero with a definite description there might be distinctions when we come to look at Kripke we'll see there are distinctions between saying that the meaning of this is given by a description as opposed to saying that the reference is fixed by a description sf0112: mm nf0098: so there might be questions about exactly what it means to say that this can be replaced by a description but to get ahead of ourselves a little bit as well it's not clear that a sense has to be a definite description sf0112: right nf0098: people like Evans kind of modern-day neo-Fregeans as they get called want to tell you that there's a notion of sense which isn't akin to being a definite description which isn't akin to say er the guy who i believe can fly and loves Lois Lane and wears a cape and all that kind of thing there's a notion of sense which is more akin to a sort of er a visual presentation or something which is kind of dependent on the object so i think it's on your handouts a bit later actually we'll see Evans has got a nice quote about just because you say that there's a way of giving an object if i give you a present there's a way in which i give you the present but that's not to say that the giving of the present is kind of indirect or that the way of giving it is something independent of the giving of the present or the present itself so he's going to want to have senses which are what he calls object-dependent they're going to be modes of presentation of an object which can't be separated from the object er that they as a sense of good sentence er but you're right to think that we might understand a sense as a definite description er and if we do then we'll see that Frege's position is going to be what's called being a descriptivist about names so we'll come on to that any other questions so far sm0113: er nf0098: namex sm0113: er does the is the problem if you see it from a use theory because then you only have pragmatic like so if you went back to the assignment of nf0098: mm sm0113: then it it seems a bit more plausible doesn't it nf0098: if you're not coming from a formal perspective sm0113: mm nf0098: yeah i think that's probably right er yeah i m-, this whole sort of problem is probably premised more on having a kind of referentialist semantics going on definitely so yeah think that's right okay er so there's a quote on the handout er i've told you that the object referred to there's two elements to the meaning of name both the object referred to and the sense sense being the way in which the name selects the object and Frege tells us in Sinn und Bedeutung that it is plausible to connect with a sign a name a word combination an expression not only the designated object which may also be called the reference of the sign but also the sense and he puts in brackets connotation meaning of the sign in which is contained the manner and context of presentation so for those guys who were paying attention when we were doing Mill last week you'll see that really what Frege's doing here is extending the notion of meaning that Mill gave us for all words for objects bar proper names Frege is extending that picture to proper names he thinks that Mill's idea that we could have names having denotation but no connotation is just wrong names have to have both sense and reference they have to have both denotation and connotation okay then well before we go on perhaps we should take a break any other questions before we go on about Frege's argument okay let's take five minutes then