Fall Quarter 2000. UCSD.
Time: Tu-Th 2:20-3:40
Room: WLH 2206
Instructor: Rick Grush (rick@mind.ucsd.edu)
Office Hours: Tu 1-2; Wed 2-3; by appointment.
Office Phone: 822-4440
[ Schedule
]
[ Quizzes and Exams ]
[ Grades ]
[ Study Questions ]
[ Exam Essay Questions ]
[ Exam and Quiz Statistics ]
[ Web Resources ]
[ Email list ]
[ How to do well in this course ]
Short Description:
Content. In this course we will explore central issues in 20th century analytic philosophy of language. The course will be broken into three units of roughly 3 weeks each. In the first we will examine Frege, Russell, and some views of the nature of linguistic representation, notably psychologism, against which they can be seen as reacting. In the second we will examine proposals concerning the semantics of demonstratives. Finally, in the third section, we will look at proposals concerning the semantics of proper names and natural kind terms.
Structure. We will meet two times a week. Each week will have a number of assigned readings, which are to be read before class. I will post study questions for the readings each week on this web page. The questions are meant to help you focus on the important issues in the readings and to help you to understand the material. There will be brief quizzes at the beginning of class as indicated on the schedule which will be easy to ace if the readings have been done, and teh study questions thought about. There will be three in-class exams. Each will consist of 10 multiple-choice (or possibly short-answer) questions and 2 essay questions. The essay questions themselves will be made available at least a week before the exam. There will be no final exam, and no paper.
Schedule:
Session 01
(09.21): Introductory Lecture.
Session 02 (09.26): [ Study Questions Here ]
Quiz 1 (on readings for sessions 02 and 03) [ Answers to Quiz 1 ]
Grush, Introductory Notes.
Locke, Selections from Essay Concerning Human Understanding
Berkeley, Selections from Principles of Human Knowledge
Session 03 (09.28): [ Study Questions Here ]
Frege, Selections from 'On Sense and Reference' (up to the middle of p. 31 in the Moore collection)
Frege, 'Letter to Jourdain' (in Moore collection)
Session 04 (10.03): [ Study Questions Here ]
Quiz 2 (on readings for sessions 04 and 05) [ Answers to Quiz 2 ]
Russell, 'Definite Descriptions' (in Moore collection)
Russell, Chapter 5 of The Problems of Philosophy
Session 05 (10.05): [ Study Questions Here ]
Donnellan, 'Reference and Definite Descriptions'
Session 06 (10.10):
Continuing discussion of Russell and Donnellan.
Session 07 (10.12): Exam 1. [ Exam 1 answers here ]
Session 08 (10.17): [ Study Questions Here ]
Quiz 3 (on readings for sessions 08 and 09) [ Answers to Quiz 3 ]
Searle, 'Proper Names'
Session 09 (10.19): [ Study Questions Here ]
Kripke, Selections from Naming and Necessity
Session 10 (10.24): [ Study Questions Here ]
Quiz 4 (on readings for sessions 10 and 11)
Evans, 'The Causal Theory of Names' (in Moore collection) [ Notes on this reading are here ]
Session 11 (10.26): [ Study Questions Here ]
Putnam, Meaning and Reference (in Moore collection)
Session 12 (10.31):
Quiz 5 (on readings for sessions 12 and 13) [ Answers to Quiz 5 ]
Evans, Chapter 11 of Varieties of Reference
Session 13 (11.02):
Readings TBA
Session 14 (11.07): Exam #2 [ Exam 2 answers here ]
Session 15 (11.09):
Langacker, 'Nouns and Verbs'. We will dicsuss only sections 3.1 - 3.5 today. This is the section that is under the link N&Va.
[This is a large file, with a lot of big images. If youand your browser are worthy, you can get the entire file here N&V. Otherwise, I have broken it down in to three more manageable chunks here -- N&Va - N&Vb - N&Vc.]
Session 16 (11.14): Quiz 6 (on readings for sessions 15, 16 and 17) [ Quiz 6 answers here ]
Continuing discussion of Langacker. (Sections 3.6 - 3.8: N&Vb)
Session 17 (11.16):
Continuing discussion of Langacker. (Sections 3.9 - 3.12: N&Vc)
Session 18 (11.21): Quiz 7 (on readings for sessions 18 and 19) [ Quiz 7 answers here ]
Grush, Brief Introductory notes on Construction Grammar
Goldberg, 'The Ditrasitive Construction'
THANKSGIVING HOLIDAY (11.23)
Session 19 (11.28):
Goldberg, 'The Caused-Motion Contruction'
Session 20 (11.30): Exam 3 [ Exam 3 answers here ]
Quizzes and Exams:
Quizzes will be administered during the first five minutes of class on the days indicated on the schedule, so don't be late (timing will be according to my watch, which I synchronize with the Naval Observatory Atomic Clock, in case anyone is curious). They will cover the readings indicated ion teh schedule. There will be no make-ups on the quizzes. I will throw out everyone's two lowest quiz scores, and so if you miss a quiz, the missed one will be thrown out. Each quiz will consist of 10 multiple choice questions, which should be easy for anyone who has done the readings and given the material minimal thought.
Exams will consist of 10 multiple choice questions, each worth 3 points, and 2 essay questions, each worth 35 points, for a total of 100 points. The multiple choice questions on the exams will be somewhat more difficult than those on the quizzes.
The essay questions will work in the following way. A week before the exam, I will hand out in class, and post on this website, a list of 5 or 6 essay questions. I will pick three of these questions to put on the exam, and you will choose any two of the three to answer. The topics for the study questions will be drawn from the readings and lectures, but it is a good bet that if you master the study questions for the readings, you will know most of what you need to know for the essay questions.
Grades:
Final Grades will be based exclusively on the exams and the quizzes. The three exams are each worth 100 points, and the six quizzes (seven total quizzes, minus the lowest score which will be thrown out) are each worth 16.7 points, for a total of 400 points for the semester. I will not assign letter grades until all the points are in. Then I will do a histogram and some statistics, and decide where to place the letter grade boundaries. But as a matter of historical record, the boundaries in the past have not been too far from a standard 10% per letter-grade scale. I will post score statistics and answers to multiple choice questions on this website after the quizzes and exams.
Study Questions:
All of the required readings are listed below, in order. For each reading, a number of questions are listed to help you focus your attention on what is important in each reading. You should be able to answer the questions, or at least most of them. Items on the quizzes will be drawn in large part from these questions. The study questions for each reading will be up at least a week before the date on which their associated lecture will occur.
Grush, Introductory
Notes.
Locke, Selections from Essay Concerning Human Understanding
(question 1)
Berkeley, Selections
from Principles of Human Knowledge (questions
2-3)
1. Explain Locke's treatment of abstract ideas. Are abstract ideas different kinds of ideas from 'particular' ideas, or are they the same kind of idea, but put to a different use?
2. In PHK(I):6-10 Berkeley considers and attacks the notion of abstract ideas. In section 10, he distinguishes three senses of 'abstraction' and says that in one sense it is a legitimate ability, but in another two senses it is not. Explain and give examples of all three senses. Compare what Berkeley says here about the OK and not-OK senses of abstraction with what Locke says about abstraction at 2:11:9. Is Locke's sense of 'abstraction' the OK sense, or one of the not-OK senses?
3. In PHK(I):11-12 Berkeley develops a theory of 'general ideas'. Why does he do this? What is the relation between 'general' and 'abstract' ideas? What, in brief, is his theory of 'general' ideas?
Session 03 (09.28.00)
Frege, Selections from 'On Sense and Reference' (questions 4-11)
Frege, 'Letter to Jourdain' (questions 12-15)
4. Frege begins with a question about equality statements, such as 'Afla is Ateb' (this example, concerning two names for the same mountain, is from the Letter to Jourdain). The question is: is this a statement about the words 'Afla' and 'Ateb', or is it a statement about the things named by those words? Frege gives an argument that it can't be about the things. What is this argument?5. What is Frege trying to illustrate with the example, at the top of page 24, of the three lines that intersect at a single point?
6. Frege introduces and discusses the technical expressions 'mode of presentation', 'sense', and 'reference' on page 24. Be sure you understand the relationship between these. With these distinctions made, what is the answer to the original question about equality statements? (That is, what information are we being given by the statement 'Afla is Ateb'? What is that statement about?)
7. At the top of page 25, Frege discusses the relationship between a sign (or word), a sense, and a reference. He contrasts what he thinks the ideal situation is concerning these relationships, but also says that natural language does not always meet this ideal. What is the ideal? There are two ways mentioned in which natural language falls short of this ideal-- one concerns the relation between sign and sense, the second concerns the relation between sense and reference. What are these two deficiencies? (Note: Ignore the discussion at the bottom of page 25 abut customary and indirect reference, etc.)
8. From the very bottom of page 25, and continuing on to page 26, Frege discusses ideas (in the sense that Locke and Berkeley understood them, as mental impressions, or qualia, or images, etc.). While Locke and Berkeley thought that words got their meaning by being associated with such ideas, Frege here argues that this can't be right. He wants to argue that what he means by the 'sense' of an expression is not the same thing as the ideas that a language user associates with it. What is his argument? (Hint: the argument concludes with the key premise of the argument, that 'mankind has a common store of thoughts which is transmitted from one generation to another.') Be sure you understand the point of the telescope analogy.
9.At page 28 Frege turns the discussion from proper names to sentences. The idea will be to see if the same distinctions and theory that he has developed to deal with the semantics of proper names will be applicable to entire declarative sentences as well. What two things would have to be associated with a sentence in order for the same theory to apply to sentences as applied to proper names?
10. On page 28 Frege provides an argument that the thought expressed by a sentence cannot be the sentence's reference. Try to understand the structure of this argument.
11. From the middle of page 28 to the top of page 30, Frege produces an argument of sorts to the effect that the reference of a sentence (if it has one) is its truth value. The basic idea behind the argument is this. First, we assume that the semantic value of a sentence is determined by the corresponding semantic value of the expressions in the sentence (we will return to this assumption in the next question). So we arrive at the sense of a sentence (the thought) by constructing the senses of the sub-sentential expressions in the right way. Now what do we get, at the level of the sentence, by putting together the references of the sub sentential constituents? This is not obvious. So Frege approaches this question by asking, what goes wrong at the level of the sentence, if one of the constituents lacks a reference? Whatever it is that goes wrong will be the thing that is normally constructed aright , from the references of the constituents, if they have a reference. What goes wrong is that such sentences end up being neither True nor False. So, this, their truth value, must be regarded as a sentence's reference. (Note: This study question is not so much a question as a brief outline of the idea behind Frege's argument. Make sure you understand his argument as he presents it, though.)
12. The first part of the Letter to Jourdain is an argument to the effect that the sense of a sentence is built up from the senses of the sentences parts. What is the argument?
13. Frege goes on to provide another argument (other than the one that was the topic of study question #4) to the effect that the sense must be distinguished from the object. What is his argument?
14. Bonus question: the names 'Aphla' and 'Ateb' were made up by Frege. What/where do these names come from? (Hint: in German, which is the language Frege wrote in, Fs are used, rather than PHs.)
15. The rest of the letter contains new versions of two arguments already given in 'On Sense and Reference'. What are they?
Session 04 (10.03.00)
Russell, 'Definite Descriptions' (questions 16-20)
Russell, Chapter 5 of The Problems of Philosophy (questions 21-23)
16. Russell begins by discussing indefinite descriptions, such as 'a man'. Definite descriptions will be more important, but it is good to get clear on the indefinite descriptions. He gives an argument to the effect that the expression 'I met a man' does not mean the same as 'I met Jones' even if, in uttering the first sentence, one was talking about a Jones-meeting incident. What is Russell's argument? How does it compare to Frege's argument that 'a = a' and 'a = b' differ in meaning?
17. Russell goes on to discuss statements involving non-existent entities, like unicorns. What challenge do such expressions as 'unicorns' present for a Fregean semantic theory -- e.g. in sentences such as "Unicorns have horns"? (Your answer should not only mention the part played by the reference and the sense of such an expression, but also the relation between them if "senses are modes of presentation of the reference" are taken at face value.)
18. Russell then targets a claim made by Meinong, to the effect that there are unreal objects (such as unicorns), which serve as the reference of expressions such as 'unicorns' (such objects, though they don't exists, according to Meinong, do subsist, whatever the hell that was supposed to mean...). Keeping your answer to the last study question in mind, how might this Meinongian idea serve to ease the tension caused to Frege's theory by 'empty' expressions?
19. Russell's aim is to show how one can provide a semantic theory that i) does not appeal to unreal objects, but also ii) can say why sentences with empty singular terms for subjects can be meaningful. This account starts at the bottom of page 48. Here is what you need to know to understand his view: Normally, we think of a simple sentence as consisting of a name, and a predicate, such as 'Agent Smith is angry.' This sentence has a name, which (simply) refers to an object (Agent Smith), and a predicate, which (more or less) refers to the property of being angry. The sentence as a whole attributes the property to the object named. Russell's point will be that in many cases, what shows up in the subject position of a sentence is not a name, but rather a description, which is basically a collection of predicates (or sometimes a collection of predicates together with some names). Russell uses Greek letters as symbols for predicates, so one way of symbolizing these sentences would be as 'The j is y'. This can be read as 'The phi is psi', as is, e.g.'the red thing is heavy'. If this is right, then so long as we have an account of the semantics of predicate expressions (which Frege will need anyway, as predicates show up in predicate position of atomic sentences), we will be able to give an account of the semantics of many of the things that show up in the subject position of sentences but do not refer to anything, things which would be an embarrassment for the bare Fregean theory, or might lead one to posit unreal objects, such as Meinong. Given all this, compare the following two sentences: "Sir Scott is angry" and "The author of Waverly is angry." Assuming that 'Sir Scott' is a name (which as we shall see in the next reading is an assumption that Russell thinks is false), how do these sentences differ in terms of their semantics, and in particular how their semantic import is constructed? Compare this to what Frege would say about the two sentences.
20. What makes definite descriptions and names seem similar is that each points out at most one thing -- each seems to refer to just one thing, if it refers at all. With names this is no big mystery, as each object is assigned a unique name, at least in ideal circumstances. But in the case of descriptions, which employ predicates, why can't more than one thing be such that the predicates are true of it? Well, for some descriptions, this will be the case. But Russell is keen to show how one can make a definite description, which will be true of at most on thing, but maybe nothing. Be sure you understand how this works (this explanation begins on the bottom of page 52).
21. In 'Knowledge by acquaintance and knowledge by description' Russell discusses two types of epistemic relation that one can have to an object (two ways that one can know about, or have mental access to, an object). It will turn out that for Russell, these two relations underlie the ways in which the two sorts of subject expressions (names and descriptions) relate to objects, and hence have semantic import. What are the two kinds of epistemic relation that Russell distinguishes? Give examples of each. Make sure you understand them.
22. Russell says "... all knowledge of truths, as we shall show, demands acquaintance with things which are of an essentially different character from sense-data, the things which are sometimes called 'abstract ideas', but which we shall call 'universals'." Universals will, to speak loosely, be things expressed by predicates, such as 'greenness' or 'tallness'. Knowledge of truths takes the form of subsuming a particular under a universal, such as 'a is green'. In this sentence, 'a' is a name, and '...is green' is a predicate. (This is not a question, so much as a note.)
23. Russell goes on to discuss what are commonly called 'proper names', such as 'Bismarck'. What does he say about them? How do these names get their semantic import? (That is, in virtue of what do they have meaning?)
Session 05 (10.05.00)
Donnellan, 'Reference and Definite Descriptions' (questions
24-29)
[Note that Donnellan
makes many references to an article by Strawson, 'On referring',
which we will not read. It is, however, in the Moore collection,
and you are welcome to read it if you like. Note also that for
electronic texts, I follow the convention of placing the page
number at the bottom of the page -- for instance, the first
page of the Donnellan article is page 361. Note also, we will
not be too concerned with anything past page 370 of this article.]
24. Donnellan's point is to show that definite descriptions, as they occur in the subject position of sentences, can work in two different ways: they can either refer, or they can denote. An expression refers to x if, so to speak, the expressions is just attached to that object like a handle. Just as, by grabbing the handle, you thereby grasp the object it is attached to, so in using a referring expression, you are thereby talking about the object it refers to. On the other hand, an expression denotes x if, so to speak, the expression acts as a filter that at most one object can slip through, and the object that happens to slip though is x. That object x, if there is one, will be the expression's denotation. For example, compare "Paul Churchland" and "the tallest faculty member in the UCSD philosophy department". The first refers to Paul Churchland, the second denotes Paul Churchland. Notice that if all of a sudden UCSD philosophy hired Reggie Miller, the second phrase would then denote not Paul Churchland but Reggie Miller. The first, however, would still refer to Paul Churchland. A referring expression, on this view, is just attached to the object, not because of any properties the object has. It is like a handle attached to just that object. A denoting expression is not attached to any object in particular -- rather, it fits just one object. But which object that is might change, or there might be no object it fits. OK, with those preliminaries in mind, and before we get to Donnellan's proposal, answer this: Does Russell recognize this difference? Approach the question this way -- Russell distinguishes names from definite descriptions; do names refer or denote on Russell's view, or does he not have a stand on this? Do definite descriptions refer or denote on Russell's view, or does he not have a stand on this? And what is Frege's position on this question concerning what he calls 'proper names'?
25. In Section III, Donnellan distinguishes two uses of definite descriptions (henceforth 'dds'): the attributive use and the referential use. Using the names of these uses as a clue, figure out which use, if either, would render the dd a referring expression, and which use, if either, would render the dd a denoting expression.
26. Donnellan produces an argument (middle of page 365) to the effect that a dd can have either of these uses. Make sure you understand the argument. On which use, if either, could I use an expression such as 'The phi' to talk about something that is not, in fact, phi?
27. Consider the following example (this sort of example is discussed around the middle of page 367 of Donnellan): You and I are at the annual MTV Philosopher of the Year Awards Ceremony in New York. A man (whose name is 'Milli') walks by us wearing a fake Rollex. Unbeknownst to me but known to you, there is another philosopher (named 'Vannilli'), backstage who is wearing a real Rollex. When Milli walks by I look at him and ask you "Who's the philosopher wearing the Rollex?" And you answer "Vannilli." According to Donnellan, what is happening in this example? That is, what kind of expression am I using, and how am I (most likely) using it, and how are you assuming that I am using it in making the answer you make? On which use, if either, have you said something true? False?
28. What point is Donnellan trying to make with the example, on page 369, involving the sentence "Is the king in his courtinghouse?"
29. This will test whether or not you really follow what's happened so far. Suppose that, in the example above, I was using the expression 'The philosopher wearing the Rollex' referentially. Would that expression then, as I used it, refer the same thing it denoted? What would the answer to this question be if I used the expression attributively?
Session 06 (10.10.00)
Readings: None
Session 08 (10.17.00)
Readings:Searle, 'Proper Names' (questions 30-36)
30. In the very opening paragraph, Searle limits his discussion to a small subclass of the types of cases that Frege argued about in 'On Sense and Reference'. He claims, in effect, that while Frege's argument in support of sense seem OK for some sorts of expressions, they don't seem OK for another sort, or so Searle will argue. What is the sort Searle will focus on, and what will he leave aside? And what is Searle's understanding of he sort he will focus on? I.e., is it the same as Russell's?
31. Searle gives an argument to the effect that, initial appearances notwithstanding, the expressions 'Cicero = Tully" and 'Tully = Tully' are both analytic. Try to understand how this argument works. It is short and manageable if you think about it. A bit later on he will say that some statements of identity, such as 'Shakespeare = Bacon' are not analytic. He doesn't explain this, and given his argument here is seems that he can't say this. Try to figure out HOW, given what Searle has said so far, he might claim that some identity statements are analytic, while others might not be (n.b. if a statement is analytic, it is not synthetic and vice versa). That is, his argument about 'Cicero = Tully' rests on an assumption about how these proper names work -- what they do and what, more specifically, they DON'T do. He might think that this is different in the case of 'Shakespeare' or 'Bacon'.
32. On the paragraph that starts at the bottom of page 167, and the following paragraph, Searle is discussing whether or not proper names have a sense. But in his discussion, he simply equates 'has a sense' with something else. What does Searle seem to think that having a sense amounts to? If this is what Frege had meant by a sense, would his account have had a difficult with empty names? Why or why not?
33. The first full paragraph on page 169 is confusing, because Searle is using terminology differently than the way we are. The following is the offending paragraph, with replacements in brackets for translating his use of terms into our use:
Let us summarize the two conflicting views under consideration: the first asserts that proper names have essentially a reference but not a sense -- proper names denote [he means: refer] but do not connote [he means: denote]; the second asserts that they have essentially a sense and only contingently a reference -- they refer only on the condition that one and only one object satisfies their sense.
34. The second full paragraph on page 169 sets up the gist of his article: it exhibits two competing views of proper names, and the rest of the article will be an attempt to decide which is right. What are the two views?
35. From the bottom of page 170 to the top of 172, Searle outlines his account. What is it? How does it differ from Russell's views about ordinary proper names?
36. When you get to the end of the article, you should be in a position to return to Searle's claim about 'Shakespeare = Bacon' and see why he thinks it is different than 'Cicero = Tully'.
Session 9 (10.19.00)
Kripke, Selections from Naming and Necessity (questions 37 - 41)
37. Kripke starts out with a set of 6 theses about the meaning of proper names which are more or less his reconstruction of Searle's account. Go through the six, and make sure you can see why each of them is something that Searle did or would need to say. [Page 72 and the top of 73 are Kripke's argument against a particular view of the meaning of proper names. We will not be concerned with this argument.]
38. On page 74 (the paragraph beginning "To summarize...") Kripke sets up the problem he will address, which is that he will argue against what he takes Searle's position to be. At this point, try to draw a comparison between on the one hand Kripke's reconstruction of Searle's view and what Kripke things is wrong with it, and on the other hand the distinction Donnellan makes between referential and attributive uses of dds. Can you see any similarities? [You can skip everything from the bottom of page 75, when he starts talking about yard sticks, to the top of page 78 at "...situation described."]
39. Kripke, starting on page 80, gives two arguments against thesis 2. What are they? [Note: Kripke uses the expression 'rigid designation' to mean pretty much what we have been using the expression 'refer' to mean -- i.e. 'expression E rigidly designates/refers to x' means something like 'E is just linked to x, not because of any properties x has or anything like that, but E is just attached to x like a handle'.]
40. Starting on the middle of page 83, Kripke turns to an argument against thesis 3. Compare this argument to Donnellan's case of 'The murder of Smith is insane.' What comparisons can you make? [You can skip everything from the middle of page 87, at "Thesis 5..." to the middle of page 91 at "...did such and such.' ".)
41. Kripke gives a brief synopsis of his account at the middle of page 91 (at "But that's not...".) and starts a more complete version at the top of page 96 (at "A rough statement ...".). Try to understand it, and then answer the following question: Return to Searle's example of the identity statement "Shakespeare = Bacon". Let us suppose that in fact, Bacon wrote all the relevant plays -- Hamlet, MacBeth, etc. Would Searle say that the identity statement was true? Would Kripke? [You can skip everything from the middle of page 97, at "I think the next topic..." to the end of the article.]
Session 10 (10.24.00)
Evans, 'The Causal Theory of Names' (in Moore collection) (questions 42 - 56)
42. Evans distinguishes, beginning at the bottom of page 208 and continuing on page 209, between two similar but distinct description theories. What are their similarities, and what are their differences? Which of these was Searle arguing for? (N.b. Evans uses the expression 'denote' to mean roughly 'refer to' or simply 'mean' -- he is not using it in the strict sense we have been using it in, to mean a relation between a description and whatever fits that description.)
43. Spanning page 208 and 209 is a paragraph marked 'b', which is Evans' recounting of Kripke's theory. The paragraph ends with this parenthetical remark: "(If it is agreed that the speaker does not denote Schmidt the conditions aren't sufficient; if it is also agreed that he denotes Gödel, again they are not necessary.)". What conditions? Necessary or sufficient for what? Understand what he is saying here and why it is true.
44. At about the middle of page 210, Evans distinguishes what he calls 'two strands' in the weak description theory. First, what is the weak description theory? The first strand basically says that a speaker must have certain intentions, and the second strand (the 'Philosophy of Mind' strand) says that there is only one way to have intentions of this sort.
45. Regarding the first strand, Evans argues that it is plausible that certain kinds of intentions are necessary in order to use language meaningfully. What is the argument for this? (Hint: it is very short, and starts with "We are prone..."). But he says that the argument often is used to try to establish that certain very specific intentions must be had, and this is what Evans disagrees with. What is the stricter sort of intention that some have argued are needed, and what is the looser requirement that Evans' thinks is sufficient?
46. The 'Philosophy of Mind' strand (which Evans will reject as to stringent) discussion is a bit difficult, but the idea is simple. It is the idea that in order for some subject S to believe that object a is F: (now I will paraphrase the logical formula) there must be some property phi such that the subject believes that exactly one thing is phi and also believes that thing is also F, plus it must turn out that a really is the one and only phi. This is more or less Russell's proposal concerning knowledge by description. For example, in order for me to believe that the cat is on the mat, there has to be some description (this will be my description that allows me to know the cat be description) in my head of some entity, and I need to believe that the thing so described is on the mat. And furthermore, my description must really pick out the cat, and only the cat. Or to put it in even simpler terms: in order to think of, or have a belief about, anything, I must have some descriptive content in my head that I believe picks out only one thing, and that description must in fact pick out the object in question. If it picks out some other object, then my belief will be about that object, and not the one in question. (Note Evans' remark near the bottom of page 211 about the meaningfulness of thought when there is no object. You can see that it is no accident that this proposal has so many similarities to Russell's view on definite descriptions -- it comes out of it.)
47. The first full paragraph on page 212 gives a brief synopsis of what Evans thinks is wrong with the 'Philosophy of Mind' (aka Russellian) account just discussed. What is it? Consider this extension of the twin example -- let's call the twins Cookie and Candy (when I was in Jr. High there were three sisters whose names were 'Cookie', 'Candy' and 'Ginger', and their surname was 'Brown'): I have a photograph which has an image that exactly matches Cookie and Candy. Could anyone (God?) tell just by looking at the picture which one it was of? The picture is, however, a picture of one of them and not the other. What makes the picture a picture of Candy, even though from the looks of it it could equally well be a picture of Cookie? Suppose that Cookie has a small mole on her cheek, and that the picture, because of a defect in the film, has a small brown spot right where the image of the cheek is. So now, the picture actually resembles Cookie more than Candy, but still, it is a picture of Candy. How can this be so? (Your previous answer should work here as well.) You should be able to see how a 'Description theory' of picture reference and a sort of 'Kripke' theory of picture reference would diverge here, and how Kripke's seems to accord better with our intuitions.
48. On page 212, Evans tests the Kripke theory against an example concerning 'Louis'. What is Evans' argument here? That is, what response should Kripke's theory force about the belief that 'Louis the XIV was a basketball player', and what does Evans say about that? What is the difference, exactly, between 'context' (what Evans appeals to) and causal links (what Kripke appeals to)?
49. At the bottom of page 214, Evans says that the causal theory (Kripke's theory) might be thought to help with problems of ambiguity. Can you make out what Evans is talking about here, and how it is that the causal account might be able to do this?
50. At the bottom of page 215 Evans comes up with examples which seem to count against Kripke's account, involving Wagera Indians and US street names. Evans does not spell out any example in detail, but you should be able to. Why would this be a problem for Kripke's account?
51. Section 3 (page 216) is where Evans turns from proposals concerning 'speaker's' meaning of a word to the different consideration of a word's meaning for a community. Evans starts with some remarks about the semantics of general terms, Forget about the fact that the remarks are about general terms, and focus on the details of the account, and its separation of 'how a word gets its meaning' and 'how its meaning is preserved'. What are these two processes on Kripke's account?
52. Consider the Madagascar example. What would Kripke's theory say about 'Madagascar' now? Ditto for the example of the switched babies.
53. Page 218 sums up the lessons so far, and looks ahead. The basic idea is that what the description theorists got right is that reference is somehow connected to the body of information that speakers have associated with the word. Kripke's account denies this. The relation that the description theorists focus on between the body of information and the referent is fit. This is what Evans thinks is wrong. Evans thinks that the right relation is something like 'causal source'. Kripke was right that causation is crucial, but Evans says that he mislocated the important causation in the initial baptism and the 'reference preserving links'. Rather, the causation that is important is that the object be the causal source of the body of information about the object. These ideas will be sharpened in Part II of the paper.
54. Evans begins discussing information, and in particular how objects can be sources of information. The idea is that each of us has a sort of mental dossier for the objects we think and talk about, and we add information to these dossiers, some of which is good information, but some of which might be misinformation. What is initially important here is that typically there will be one object which is the dominant source of information in a given dossier. Evans also discusses 'dominance' of information, which seems to be something like how important a piece of information is for a speaker. Be sure you understand the Napoleon example.
55. Page 222 Evans provides a brief synopsis of his view (points 1 and 2 on that page). The main example here to try to understand is the 'Turnip' example at the end of the article. The examples in between page 222 and this last example are all concerning special cases, cases of deferential reference, etc. You can safely not worry too much about them (but do read them). The main example, which Evans takes to display the theory as a whole is the 'Turnip' case.
56. What is the difference between the old villagers and the young villagers in the 'Turnip' example? What would Kripke's theory say about this case? What does Evans' say? Be sure you can tell WHY it says this. Specifically, what is the process by which names get their meaning (see study question 51), and when and how is it operative in this example?
Session 11 (10.26.00)
Putnam, 'Meaning and Reference' (in Moore collection) (questions 57 - 61)
57. Putnam begins with a barrage of terminology that we have not encountered (this terminology is due to Carnap, whom Putnam will mention but not focus on). Don't worry about it. By 'extension' he means roughly 'reference', the actual stuff that the term refers to. By 'intension' he means, roughly 'sense'. The abstract entities Putnam attributes to Frege are of course senses. Get used to reading Putnam's 'extension' as the same as 'reference', and 'intension' as the same as 'sense'. This replacement should begin at or even before the second paragraph. So Putnam says in the second paragraph that while it is common knowledge that two terms can have the same reference but different senses, the reverse is usually held to be false -- you can't have two terms with the same sense but different references.
58. On page 152 Putnam discusses how the Twin Earth case does not create a problem for the meaning of the word 'water', saying that it has two meanings. This is something like how 'jumper' means 'sweater' in the UK, but 'person intending to commit suicide by plummeting from a high place' in the US. Anyway, be sure you understand the argument that Putnam gives, which ends with his conclusion that the reference of the term water is not a function of the psychological state of the language user. Note: the term 'idiolect' has shown up both in this article and in Evans'. A dialect is a special form of a language spoken by a smaller community of people. An 'idiolect' is a special form of a language spoken by one person -- it is an idiosyncratic dialect. Dig?
59. Compare Putnam's conclusion that 'meanings just ain't in the head' to Kripke's and Evans' proposals about the meanings of proper names (as compared to, for instance, the proposals of Russell and Searle). What obvious comparison can be made?
60. A definite description is a set of properties that specify necessary and sufficient conditions for something to be the denotation of an expression -- it is the denotation if and only if it, and only it, satisfies all of the properties as specified in the description. Given this, Russell's proposal is that one knows the meaning of an expression if one knows the definite description it corresponds to -- that is, if one knows the necessary and sufficient conditions for being the reference of the expression. Compare this to Putnam's proposal for the meaning of the term 'gold'. There is a sense in which Putnam agrees with the Russellian idea, but a very important sense in which he disagrees. What are they?
61. Don't bother with the section 'Indexicality and Rigidity' which starts on page 157.
Exam 1.
Essay Questions:
1. In 'On sense and reference' Frege considers two plausible accounts of what the meaning of a singular expression (e.g. "Ateb") consists in, and he rejects them both. Identify the two accounts he argues against, and explain his arguments against them.
2. In 'On sense and reference' Frege offers a positive account of what is needed, in addition to an appeal to a term's reference, to provide an adequate account of the semantics of proper names. What is this additional thing that each term has? How does Frege characterize it? What are its properties? What is its relation to the contents of individual minds? What is its relation to the term's reference?
3. If the sense of an expression is the 'mode of presentation of the reference', then it would seem that expressions which lack a reference would also lack a sense -- there has to be a reference in order for it to be presented in some way or other. This seems to have the consequence that 'empty' expressions are meaningless, yet clearly many such expressions are perfectly meaningful. Russell tries to show how some expressions which fail to refer can nonetheless be meaningful. Clearly explain how he does this. Your answer should include (but not be limited to) discussion of the distinction between names and descriptions, and the two types of knowledge Russell claims one can have about an object.
4. Discuss Russell's view of names. Specifically, what does Russell think about normal proper names in natural language (such as 'Bismarck' or 'Plato')? Are they really names (in Russell's technical sense)? What are logically proper names? What sorts of things can logically proper names refer to? What is the relationship between a normal proper name, such as 'Plato', and definite descriptions? What is the relationship between logically proper names and definite descriptions?
5. You and I are at the annual MTV Philosopher of the Year Awards Ceremony in New York. A man (whose name is 'Milli') walks by us wearing a fake Rollex. Unbeknownst to me but known to you, there is another philosopher (named 'Vannilli'), backstage who is wearing a real Rollex. When Milli walks by I look at him and ask you "Who's the philosopher wearing the Rollex?" And you answer "Vannilli." According to Donnellan, what is happening in this example? That is, what kind of expression am I using, and how am I (most likely) using it, and how are you assuming that I am using it in making the answer you make? On which use, if either, have you said something true? False?
Exam 2.
Essay Questions:
1. Searle discerns an unacceptable consequence of a 'Russellian' theory of proper names (that is, a theory according to which a proper name is just a disguised definite description). What is this consequence and why does it follow from the 'Russellian' view? Explain Searle's own theory, and how it avoids the problem that the 'Russellian' theory faced.
2. Draw a comparison between i) Kripke's reconstruction of Searle's view and what Kripke things is wrong with it; and ii) the distinction Donnellan makes between referential and attributive uses of dds and his arguments for the relevance of the distinction.
3. Russell's theory of descriptions was motivated by what Evans (in material we have not read) calls Russell's Principle, which is: In order for a speaker to refer to something, that speaker must be able to distinguish that thing from all other things. Putnam's theory rejects one strand of Russell's Principle, but keeps another strand. Elaborate. That is, explain Putnams' theory in a way that makes it clear which aspect of Russell's Principle it rejects and which it accepts.
4. Evans' theory, as developed in 'The Causal Theory of Names' and 'Proper Names', has similarities both with Searle's 'cluster of descriptions' theory, and Kripke's theory. But it also differs from both. Using Section 6 of 'The Causal Theory of Names' (where Evans himself remarks briefly on the similarities and differences) as a starting point, (ignore the sentence about 'logical doctrines') explain and elaborate on the similarities and differences.
5. Searle, Kripke and Evans give us 3 different theories as to how the reference of a proper name is established. Construct a single example (something roughly in the style of Evans' "Turnip" example, or the example involving X, Y and Z that was discussed in class on Thursday) in which we have a single name, "Felony", such that because of the origin of the name and the sources and details of the information circulating with that name, the name would refer to one of three different people depending on which theory was correct. That is, develop an example in which there is a community of speakers who use the name 'Felony', and is such that Searle's theory would say that 'Felony' refers to A, Kripke's theory would say that 'Felony' refers to B, and Evans' theory would say that 'Felony' refers to C (where A, B and C are all different people).
Exam 3.
Essay Questions.
1. Using the example After I ran over the cat, there was cat all over the driveway as the primary test case, explain Langacker's account of the mass/count noun distinction. Your answer should include (but not be limited to) discussion of the criteria for something being a mass or count noun, and why these criteria are met in the two instances of 'cat' in the example.
2. Langacker claims to give an explanation for why the only class (excluding special contexts such as announcing and story-telling) of perfective verbs to appear in the simple present in English is the class of explicit performatives, e.g. I order you to put down that rifle; I pronounce you man and wife. Explain this argument. Your explanation should include discussion of the semantics of present tense in English, and a discussion of the contrast in the relevant semantics between perfective and imperfective verbs.
3. Langacker claims to provide an explanation for why perfective verbs (except explicit performatives) must appear in the progressive construction, and not the simple present, in the present tense (I am building a canoe; *I build a canoe), while imperfectives pattern in the opposite way (*I am resembling my father; I resemble my father). Your answer should include a discussion of the semantics of perfective and imperfective verbs, the present tense, and the effects of the components of the progressive construction 'be' and '-ing'.
4. The ditransitive construction (SUB V OBJ1 OBJ2) and the dative construction (SUB V OBJ2 'to' OBJ1) can be used to describe many of the same events, e.g. Mary gave the baby a toy; Mary gave a toy to the baby. But the ditransitive construction has constraints which the dative does not, which explains why Mary threw the ball to the fence is OK, but Mary threw the fence the ball is not. Consider a case where I am sending a letter to my girlfriend in Europe, and I say I am sending a letter to Europe, but I couldn't say I am sending Europe a letter. But note that it would be possible to say Clinton sent China a stern letter, warning of future sanctions, where we seem to have a ditransitive clause parallel to the infelicitous I am sending Europe a letter, but in this case it is OK. Explain the relevant constraints, and why the 'Europe' sentence is bad, but the 'China' sentence is OK.
5. Consider these two examples of caused-motion clauses: a) John lured him out of the bar. b) *John persuaded him out of the bar. Goldberg argues that (b) is infelicitous because the semantics of 'persuade', unlike 'lure' conflicts with one of the semantic constraints on the Caused Motion Construction. Identify and explain this constraint, and run through the argument for why 'persuade' but not 'lure' conflicts with it.
6. Consider these two examples of caused-motion clauses: a) Mary nudged the golf ball into the hole. b) *Mary nudged the ball down the incline (to describe a scene in which the ball was resting at the edge of a large incline, and Mary's nudge started it rolling down this incline). Goldberg argues that (b) is infelicitous because a semantic constraint on the Caused Motion Construction. Identify and explain this constraint, and run through the argument for why (a) but not (b) conflicts with it.
Exam and Quiz Statistics:
Quiz average scores (out of 10 possible):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Average scores on Exam One:
|
|
|
|
|
|
|
|
|
|
|
|
Average scores on Exam Two:
|
|
|
|
|
|
|
|
|
|
|
|
Average scores on Exam Three:
|
|
|
|
|
|
|
|
|
|
|
|
Web Resources:
Here you will find the web page for a philosophy of language course being taught by Professor Pete Mandik (former student of mine, so you can be assured he knows his stuff). His class (at William Patterson University in New Jersey) is reading some of the same things we are, and as Pete has some notes and stuff up on the site, it might be worth a visit.
There is an email distribution list for this course -- 134@mind.ucsd.edu -- and subscribing to it is required. The two purposes of the list are i) to allow me to distribute information regarding changes of schedule, etc., and ii) to provide a means for students to ask questions and continue discussion outside of class time. The list is set up so that all and only people subscribed to it can post to the list.
To subscribe: send an email to listserv@mind.ucsd.edu. The subject line should be blank, and the body of the message should contain one line of the following form:
SUBSCRIBE 134 FIRSTNAME LASTNAME
Where the "FIRSTNAME" and "LASTNAME" are place holders for your actual name. For example:
SUBSCRIBE 134 Gottlob Frege
You will receive automatic confirmation that your subscription was processed, or an error message if there was some problem. [NOTE: Some email programs, notably MS Outlook Express, add formatting characters to the beginning of outgoing messages (MIME coding) that screws up the listserv program. If you receive a 'syntax error' message, but are sure you did everything correctly, this is probably the problem. To get around it, send the email from a bare bones program, like PINE or UNIX Mail. Otherwise, try to get your email program to send the message without any formatting characters.]
To post: send the message you wish to post to 134@mind.ucsd.edu.
[Note that the address via which one subscribes is different than the one to which one posts messages. Subscription requests are sent to listserv@mind.ucsd.edu and posts are sent to 134@mind.ucsd.edu]
How to do well in this course:
1. Be sure to read the assigned material before class. This will help to ensure that our class time is quality time, and not wasted with me and some of the students reiterating the material to those who haven't read it. Reading the material before class will also help you get some points on the quizzes, and that is a good thing.
2. Use the study questions to help you focus on the important topics in the readings, and also to help you determine if there is some topic you don't understand. Try to get to the point where you would feel comfortable answering all the study questions before class meets. But if there are a few that you are still finding difficult, don't panic.
3. Bring specific questions with you to lecture. Even if there are aspects of the readings or study questions you didn't fully grasp, the fact that you tried should help you to narrow down what it is that you don't get.
4. See me in office hours, or email me, if there are still questions you are unclear about. That's what office hours are for. Don't be shy.
5. After class, re-read the material for the session, paying special attention to the questions that you had before. Hopefully the material will make much more sense to you now.
6. Manage your time well. Many students are lazy, and wait until a few days before exams to start studying. That is bad time management. The same number of hours devoted to the material BEFORE lectures can lead to a much more efficient use of your time, better understanding, and higher grades.