Notes On Philosophy of Language

I recently read Philosophy of Language: A Contemporary Introduction and enjoyed it. I took notes on it in Evernote and couldn’t think of a reason not to share them, so:

  1. Introduction: Meaning and Reference
    1. It is an amazing fact that certain marks have meaning to us.
    2. A philosophical theory of meaning should account for why things have meaning and how they get their meaning.
    3. The Referential Theory
      1. The naive approach to language is that all words refer to something and that thing provides their meaning. It is easy to see how this theory falls short, as in the 4 paradoxes (e.g. The King of France is bald) as well as by examining parts of speech that have no reference (“as”, “the”, “between”, etc).
  2. Definite Descriptions
    1. A definite description is a set of words that begins with “the”, like: “the cat on the table”. The Referential Theory would have us believe that this refers to a particular thing in the world and this get their meaning from them. Bertrand Russell analyzed these sentences and came up with his Theory of Descriptions. He contended that these sentences have a subject-predicate form grammatically, but a much more complicated logical form underlying them.
    2. Russell’s Theory of Descriptions
      1. Russell contended that a definite description actually states 3 things:
      1. There is at least one X.
      2. There is at most one X
      3. X has property Y.
    3. For example, the statement “[The King of France] is bald” is really asserting:
      1. There is at least one King of France.
      2. There is at most one King of France.
      3. The King of France is bald.

        This  analysis allows us to solve the “4 paradoxes” since in this example we can see Assertion #1 is incorrect. In this way, Russell tries to assign meaning to specifically definite descriptions.
    4. Russell introduced existential quantifiers like “At least one, At most one, All, No” into the logical form of sentence structure.
    5. Objections
      1. Many people objected to Russell’s analysis – indeed, it “feels” very far removed from the way we think about things.
      2. Strawson says that no one feels “The King of France is bald” to be false. They more feel it to be nonsensical.
      3. Donnellan makes a distinction into two types of sentences: attributive and referential. He asks us to picture us standing over the dead body of Bob. We might claim “Bob’s murderer is insane!” In thisattributive sense, Russell is correct – we are making a logical claim and not referring to a known person. Imagine now we are at the trial for the murder and a person is about to be convicted. They are drooling. If we are in court and I utter: “Bob’s murderer is insane!” SURELY I am referentially talking about the person on the stand. Donnellan says that Russell is discounting this massive use of definite descriptions.
  • Proper Names: the Description Theory
      1. Russell created the Name Claim, which says that proper names are really short hand for definite descriptions. Suppose you say “I am going to see Neil Nelson.” If I don’t know the proper name, how would I respond? I would say something like “Who is Neil Nelson” and you would respond with a definite description: “Neil Nelson is the person who did x y and z.
      2. John Searle objected by saying that different people might know the same proper name by different definite descriptions and surely they don’t have the same meaning inside of their head for the relevant terms. This led Searle to come up with his Cluster Theory, where a sufficient cluster of definite descriptions uniquely defines a proper name. This helps, but the theory is still doomed.
      3. Saul Kripke comes in from a new angle by considering Possible Worlds.
  • Proper Names: Direct Reference and the Causal-Historical Theory
    1. Kripke says possible worlds are important because you can envision a possible world where “The president in 1970 wasn’t the president in 1970” but you CANNOT envision one where “Nixon isnt Nixon”, so definite descriptions CANNOT be simply reduced to proper names.
    2. Kripke goes on to claim that definite descriptions are flaccid and proper names are rigid in their connections to their meanings.
    3. Direct Reference theory says that proper names directly reference things.
    4. The Causal Historical Theory claims that there have been a historical chain of utterances of a proper name that branch out from an origin (birth) to the current state and that each utterance’s meaning has come along the way.
    5. Twin Earth – another earth just like ours with everything the same except for the (undiscovered) compositional structure of water would come up with the same terms as regular earth for everything. This causes Kripke to say “meaning ain’t in the head”
  • Traditional Theories of Meaning
    1. A theory of meaning must explain the “meaning facts”. For instance, how a word can have a meaning, how two different expressions can have the same meaning, how an expression can be ambiguous, etc.
    2. Entity Theories
      1. Ideational Theories – Starting with John Locke, these claim that meaning is all inside of the head and that utterances express a mental state.
      2. The Propositional Theory – underlying, eternal propositions are expressed by sentences. For example, the sentence “the snow is white” in two different languages are expressing the same proposition. There is something strange about this theory because the propositions don’t exist in time and space and are eternal, but this theory, like other theories, helps us shed some light on the subject (similar to subatomic particles).
  • “Use” Theories
    1. The entity theories got to part of the issue, but did nothing to really show how words could  CAUSE an action in the world – that is, they didn’t really get at the push/pull of real things that language can have.
    2. Wittgenstein believed language to be a “use”. We make an utterance to effect some change on the world. He believe Russell’s analysis methodology of putting things on a blackboard to be utterly incorrect. and abstract.
    3. W believed that language uses were a series of language games – that is, similar to games like chess, language was used with certain rules in certain contexts. You learn the rules of how to play and then you can effectively communicate in that game. Imagine a builder and her apprentice. The builder needs a block or a slab. The builder calls “Block!” and the apprentice hands her a block. What does “block” mean here? In this language game, it means to hand a block to the builder. It does not refer to the block directly but is part of a “use”.
  • Psychological Theories: Grice’s program
    1. Grice comes along and says that when you say something, it is usually with the purpose of communicating an idea, whether by delivering an opinion, expressing a desire or intention.
    2. Grice distinguished between “utterer’s meaning” (or “speaker meaning”) and the meaning of the utterance by itself.
    3. Grice offers a two staged approach:
      1. Reduce sentence meaning to speaker meaning.
      2. Reduce speaker meaning to a psychological state.
    4. Speaker meaning: “S uttered x intending that A form the belief that S believes P. Basically, you say something in the hopes that someone else understands that you have some belief P.
    5. To overcome objections, this theory has to concede too much to other theories.
  • Verificationism
    1. A sentence is true if and only if its being true would make some difference to the world.
    2. This theory (tied with logical positivism) is part of the general movements to put everything under science’s umbrella. It has requirements for verification and falsification (including testibility). A major objection is that you would have ot understand what some utterance means before you could verify it, but now it is already too late.
    3. The verifiction principle cannot itself be verified.
  • Truth-Condition Theories: Davidson’s Program.
    • Davidson makes a claim for the compositionality of meaning in sentences. He makes this argument by pointing out that we can quickly and easily understand complex sentences that we have never seen before and further that sentences can have infinite length. If they can truly be infinite in length, we must be combining previous notions that we have to come up with meaning. He also argues that syntax is important.
    • Davidson claims that not the verification method defines meaning but the truth conditions. That is, the set of sentences that make the sentence true.
    • Using logic and grammar we can analyze the truth conditions of the subsequent parts of a complex sentence.
  • Semantic Pragmatics
    • Linguistic Pragmatics is the study of language in social contexts.
    • There are two types of pragmatics:
      • Semantic Pragmatics – Looking for contextual information INSIDE of the proposition itself. (e.g. “That needs to go.”)
      • Pragmatic Pragmatics – Context outside of the proposition (e.g. Someone points to something and says “I mean that.”)
      • Some people tried to fix the context to popular 8: world, time, speaker, hearer, etc, but this is inflexible as context can occur in many different ways.
  • Speech Acts and Illocutionary Force
    • Performative utterance is when you do an act just by speaking: “I raise”. These don’t seem to be truth functional – they just…are.
    • JL Austin tried to find a test for performatives – he came up with “hereby”. If “hereby” can be inserted before the verb then it is performative: “I hereby raise.”
    • He came to find however, that the performative/constantive distinction didn’t hold up in a lot of cases, like “I state that I have never traveled to a Communist country.”
    • Austin came to realize that most utterances have both a performative and a constantive aspect.
    • Illocutionary force is the force of the statement (the social act) that isn’t the direct content of the proposition.
    • Searle divided speech act rules into:
      • Consitutive – Rules that define the act itself (the rules of chess define chess)
      • Regulative – regulate a form of behavior and are independent of that form of behavior (etiquette rules)
    • The truth of a performative statement can be seen as the truth of the illocutionary part and the truth of the propositional content combined.
  • Implicative Relations
    • It is often the case that the utterer of a sentence imply things that underlying proposition does not entail. This happens in a lot of different ways.
    • Conveyed Meaning – a conveyed meaning is a meaning of an utterance that is meant (usually by the speaker) to contain something different than the underlying proposition. Example: To a house guest who is being rude one might say “There’s the door.” While the proposition “There’s the door” in and of itself means…well…”The door is over there.” it is clear to all parties what is implied by this statement.
    • Invited Inference – Another type of implication is Invited Inference, where humans add logical form/causality/temporality to an utterance that doesn’t contain it. For example: “If you mow my lawn, I’ll give you ten dollars.” This is a conditional, and strictly from the proposition you shouldn’t rule out that I’ll give you ten dollars regardless of whether or not you mow the lawn, but you would surely interpret it as “If and only if you mow my lawn, I’ll give you ten dollars.”, thus making it a biconditional. One can see how “A and B fell in love and they got married.” is different than “A and B got married and they fell in love.” are different, even though the underlying logical structure is the same.
    • Conversational  Implicature
      • Grice (1975) who we dealt with in the chapter on Psychological Theories, developed a theory he called Conversational Implicature. This theory attempts to explain how we handle implications in language.
        • The first part of the theory is the Cooperative Principle – this states that communication should be as much is required at the time it occurs for the purpose of the talk exchange. The interesting part are the Maxims that attach to it:
          • M1 – Make your contribution to a conversation as informative as is required (“Maxim of Strength”).
          • M2 – Do not make your contribution more informative than is required.
          • M3 – Do not say what you believe to be false (“The Maxim of Truthfulness”).
          • M4 – Do not say that for which you lack adequate evidence (“The Maxim of Evidence”).
          • M5 – Be relevant (“The Maxim of Relevance”).
          • M6 – Avoid ambiguity.
          • M7 – Be brief (avoid unnecessary prolixity).
        • With this theory, Grice says that people first recognize that someone is not observing the maxims, therefore they content must be something else. Then they search for the meaning of the utterance through a logical inquiry of the maxims. In the door example one might reason: “He said ‘There’s the door.’, but by the M5 I can see that that is not relevant in the normal context – so what is he trying to show me? He knows it is not relevant, knows I know the rules, and yet said it anyway…hmmm.” Of course this reasoning happens instantaneously and unconsciously, which is a problem that philosophers have with the theory.
        • Implicatures are cancelable – someone can make the implication null by stating that they did have the original intention of the proposition: “No, I really mean – there is the door. I wanted to show you because we repainted it last week.”
        • Although most philosophers agree with some version of this theory, 2 complaints are:
          • It doesn’t seem satisfying that extremely complex implications are figured out on the fly instantaneously. The theory has nothing to say about this.
          • The theory doesn’t explain much about the “figuring out” of the context in the “positive” part of the theory. For example, in the door example, the “negative” stage (“He isn’t talking about the door at all!”) makes sense, but the “positive” stage, finding out the new meaning, is largely unaccounted for.
      • Presupposition
        • This goes back to Strawson’s complaint with Russell about the Theory of Descriptions – that “The present King of France is bald.” is not false, but lacks truth value. This is because it contains a presupposition that is false. When a sentence entails a full proposition and the proposition is false, the sentence is false, but when it presupposes a proposition, it lacks truth value.
        • Example: “Have you supposed beating your spouse?” presupposes “You have beaten your spouse.”
      • Relevance Theory
        • A bunch of theorists (originally Sperber and Wilson 1986) began building off of the Gricean model, but then became competitors. Their claim is that humans say whatever is relevant through a combination of mental processes and are simply not aware of any Conversational Maxims. Relevance is understood as a favorable balance of “positive cognitive effects” over processing time and effort. We speak loosely sometimes, like calling a CD a record, and no one challenges the meaning because it is clear.
      • Indirect Force
        • Moods (declarative, interrogative, imperative) seem to have their place, but the picture is often much more complicated:
        • “Can you pass the salt?” is a question, but really a command, and further – it isn’t asking about the physical capability of the hearer. We disentangle these contexts instantly. Searle invokes Gricean reasoning to solve the problem. Something like “Well whether or not I can physically pass the salt is obvious and irrelevant. So…”
  • Metaphor
    • Although philosophers generally think that metaphorical expressions are the exception and literal expressions are the rule, an exploration can show this to be false. For example, consider how often a word like “level” is invoked. This used to represent something horizontal in the physical world, but now we use it metaphorically all the time.
    • Interestingly, many metaphors become “dead” – they lose their figurative meaning and take on a literal meaning. For example, the “mouth of the river” used to be a metaphor to a person’s mouth. Now no one thinks of that when the sentence is uttered.
    • A good theory of metaphor must explain: 1) What “metaphorical meaning” is, if anything and 2) By what mechanism is it conveyed?
    • A general explanation for categorizing metaphor is “With such a sentence there is a conceptual “tension” – in “Juliet is the sun.” there is a tension between Juliet and the sun, as these two things are categorically different.
    • Davidson’s Causal Theory
      • Davidson denies that there is anything linguistic going on in metaphor at all. He claims it is purely causal in that the sentence causes us to attend to some likeness, often a novel or surprising likeness, between two or more things. He believes taking a pill or getting a bump on the head could make you receive the same knowledge. Therefore, it is not in the linguist’s domain.
      • Davidson denies that there is “metaphorical [linguistic] meaning” at all (similar to the logical positivists, as they would say it lacks verification). Certainly there may be an emotional or affective significance.
      • Under this theory, no one can misinterpret a metaphor – they may just have different architecture.
      • Against this theory: When a metaphor dies it takes on a literal meaning. This would be hard to explain if there were never an original meaning to begin with.
      • Better theories that reject metaphorical sentence meaning have been proposed, so there is no reason to accept this theory.
    • The Naive Simile Theory
      • The common theory: that a metaphor is just a shortened simile.
      • Beardsley says this seems wrong, as if it were merely short for the simile it would be synonymous with the simile, but this feels wrong
    • The Figurative Simile Theory
      • Similes are often used as figures of speech. When a metaphor is used it can be seen to represent a simile that is a figure of speech, but not a literal metaphor. This solves issues raised with the Naive theory, but really defers the task of explanation to the figurative simile.
      • Fogelin says that using “like” is not just the bidirectional sharing of properties. Really, “A is like B” means that A has the salient properties of B. For example, a chipmunk is like a rat, because it has the salient properties of a rat, but a rat is not like a chipmunk, because a chipmunk’s salient properties include cuteness, which a rat does not have. Fogelin says that a figurative comparison involves the salience standard where as a literal comparison does not.
    • Pragmatic Theory
      • Searle goes back to the Gricean view of communication for implication. This means the hearer determines the speaker meaning from the context and constructs theories of the meaning. If there are many applicable ones (think abstract poetry), then the meaning is open-ended.
    • Metaphor as Analogical
      • Infinite Polysemy – words take on new, related meanings all the time from old meaning.