Over at the Valve the discussion of Derrida's On Grammatology turned into a discussion of Peirce. (OK, I was the cause - guilty as charged) I figured though that it would be helpful to understand what Peirce means by signs since it is an amazingly useful way of thinking. I think semiotics, unfortunately, is neglected in philosophy a tad too much. In my opinion it simply opens up many problems in a fashion better than how they are typically considered. So I figured I'd do a bit of a primer. I'll include links to quotes from Peirce for each of the terms.
The basic notion of a sign is "anything which is so determined by something else, called its Object, and so determines an effect upon a person, which effect I call its Interpretant, that the latter is thereby mediately determined by the former." (SS 80) In other words a sign stand for something (it's object) and creates an effect (the interpretant) Thus the sign's standing for its object produces an idea.
Now typically when we talk about signs we're talking about symbols. However Peirce divided signs into three categories: symbols, icons and indexes. The categories were based upon the relationship between the sign and its object. If the association was due to a resemblance then the sign is an icon. If the association was due to a "real connection" and not a resemblance then the sign is an index. (Typically this "real connection" is a causal one) If it is associate by neither resemblance nor a real connection it is a symbol.
Of course in practice we have degrees. So, for example, my image in a mirror is part icon since it resembles me and part index due to its causal relationship to me. A weather vane, however, points the direction of the wind primarily due to a pure indexical relationship. Words on the page are conventional and thus associate with their object as a symbol.
In addition to the kinds of signs categorized by their object we have kinds of signs categorized by the interpretant. (That "effect" - typically the creation of an idea in someone's mind) Peirce categorized these based upon the notion of a proposition with pieces missing. (See for instance this excerpt from "The Categories Defended")
We have the part of the proposition we call a predicate and when Peirce called a rheme. This is formed by removing from a proposition the name or noun. It is characterized by being neither true nor false but is a kind of pure possibility. It is the interpretant of the object considering the object only in terms of it's characters. (More than just its properties but independent of the substance of the object)
The next category of sign in terms of interpretant is a proposition or what Peirce calls a dicent. To Peirce this is roughly an interpretant that is true or false but doesn't offer a way of determining if it is true or false. It is, for Peirce, a sign of actual existence and represents its object as being existing.
The final category is argument. This one's trickier even though it seems obvious, given its name. To Peirce this is an interpretant that represents the sign-like aspects of its object rather than its existence or character. He sees this as an aspect of law which makes sense when you think of an argument as moving from premises to conclusions in a lawlike fashion.
Next up are signs in terms of their sign-like aspects. The first is a qualisign which is a sign of a particular phenomena or appearance. (A sign in terms of is presentation to me) It is roughly any quality in its sign like aspects. The second is a sinsign which is a sign as an individual object or event. (A singular) The third is a legisign which is a general type or law. While we general think of convential signs as legisigns it is broader. It signifies through an instance of its application.
All of this seems fairly complex. But I bring it up simply to highlight the nature of signs. It's easy to think of signs and then pick out just one of these aspects as if that was what fully constituted a sign.
The second thing to keep in mind is that we don't encounter signs in a vacuum. So consider the word "the." It obviously occurs many times on a single page. Each of these occurrences is the same word (a legisign) but each occurrence (instance) is a replica of this word and thus a sinsign. So many signs require other signs to function. (An argument being the obvious example)
The last aspect of signs one has to consider is Peirce's famous type-token distinction. We're all familiar with this. (Especially if one is a computer programmer) The actual instance I encounter any sign is a token. Thus the particular glyphs from a particular page spelling out "the" is a token. So a token is an other name for a sinsign. It's a particular event or thing as sign. The type is the law or general idea as a sign. So our idea (sense) of "the" is a type. Types are legisigns. In addition Peirce talked about the tone. This in a simple form is the manner in which one creates a particular sign. So I can talk about the tone of voice I use to say "the" or the style which its token on a page has.
Now I'll be the first to admit that the distinction between tone and token is blurry. There have been a fair number of debates on this. Personally I think sometimes Peirce talks about individual objects as tokens and then sometimes as tone. So he's not always careful about the phenomena vs. object distinction. However as I said, these categories come in degrees. The idea is to simply keep in mind that there are these different ways of thinking about signs so as to be aware of them. But in practice things are always a bit more blurry.
One debate is over the word for a concept as it is found in different languages. Say the English word "dog" versus the Spanish word "perro." Some would argue that these are examples of tone while others would call it a token (qualisign). And the concept dog would be the type (legisign). Not everyone agrees though. (And of course there is then the problem that often as we switch languages the types are similar but not identical)
(Note: I lifted the illustrations off of the paper "The 1903 Classification of Triadic Sign-Relations". It goes further than I do in the above and analyzes when we combine these different aspects of signs. )
I'd written this for Peirce-L that some might find helpful.
Perhaps it's just me but find that the tone/token distinction isn't held to very strongly. It would work best if token was always the phenomena or appearance whereas tone was the object with its indefinite ways of appearing. But Peirce doesn't talk that way. So he talks about a word on a page of a newspaper as a token. But because the word on the paper can appear in different ways it is closer to tone than token. (IMO)
Of course if we just acknowledge that these "ideals" never appear in their ideal form then the problem goes away. So we could consider the word on a paper as a token in an "idealized" consideration. (Ignoring the individual event of its appearance to a reader) Likewise we can consider the word on paper as a tone.
In addition we can consider the class of words on a page as a token. Roughly the word in an indefinite form as it can be instantiated in different fonts (Geneva, Times Roman, Helvetica), in different colors, be read in different lighting conditions. But we can also consider this class as a type (the abstract type of the forms of the glyph)
I suspect any particular sign will always be analyzable *simultaneously* as a sign manifest as a token, tone, and type.
I'm confused -- what is the benefit to making all of these subtle distinctions and employing all of these novel terms? What sort of philosophical muddles does this help keep us from falling into?
(reading some of Pierce's definitions that you linked to reminded me of trying to get into Hegel -- it was not helpful to be given a "definition" of, say, identity, that presumed I already knew four other Hegelian terms of art. It took a good bit of effort to find a place where I could edge into Hegel's train of thought; the "Philosophy of Spirit" ended up being the key -- I could parlay my theological background into an understanding of several of Hegel's concerns, and then eventually work backwards until I was able to make some progress on the logiky bits.)
Incidentally, since you'd mentioned on The Valve that you were trying to get into Hegel a bit more: I found Jon Stewart's "The Hegel Myths and Legends" to be very helpful. I've not heard good things about Solomon's Hegel book, but I haven't read it myself.
I think keeping signs clear is intrinsically helpful so you don't neglect some aspect of how they work. I take much of Derrida's work, especially Limited Inc. to be going through a lot of this.
Peirce and Hegel are (according to Peirce anyway) similar in some fashions, although Peirce was pretty critical of Hegel on a lot of matters and thought the Phenomenology a mess and largely worthless except for some of the categories of three Hegel came up with. Both came out of that era when taxifying things was a big part of philosophy I think. Peirce in particular attempted to learn by putting things into categories although there are obviously reasons for that given that he replaces the Kantian categories with his three categories with everything else being a variation on that.
Anyway, I'll see if I have time later this week and go through some examples from Derrida that I think Peirce illuminates.
I should add that a lot of this seems cliche simply because many of Peirce's categories have been adopted wholesale in various disciplines. (Although Morris' linguistic categories are actually subtly different from Peirce)
Rheme is the one that I still can't work out. The first and the second trichotemies are fine, but the third trichotomy gives me trouble, with the Rheme, in particular, baffling me.
I thought I was on the verge of working it out, but then I came across this at the very bottom of the linked page:
"Every verb is a rhema."
Now I'm back to being baffled.
Any chance you might elaborate?
Cheers
Rhemes are predicates separated from the subjects that are doing them. So if I say, "the ball bounces high" then "bounces high" is the rheme and is a verb or more properly a very phrase. So I think he's using verb a bit expansively in that comment. And of course I'm simplifying so as to not get into too much grammatical jargon.
Peirce characterizes rhemes are pure possibilities since in and of themselves they don't exist. (There is no Platonic existence of "running", for example) They are pure possibilities. The possibility of running can be exemplified when I run or when you run. This notion of possibility is important in various ways for Peirce's logic.
In the chart above each of the little numbers are actually important. They tell which of the three categories of Peirce are being appled. So we can conceive of the interpretant in terms of firstness (in itself without external relations - thus pure possibility or character); secondness (a two part relationship: in this case subject and predicate forming a proposition); thirdness (semiosis or in this case an argument or process of reasoning)
Thanks Clark.
I'd read the comparision to predicates a few times before, but it wasn't really helping. The link to firstness does it though.
My Peirce is actually very limited -- just a few extracts from the Collected Papers bundled together as a piece called "Logic as Semiotic" for a semiotics reader.
Cheers
I've closed comments in order to avoid spam since I don't check this older blog as much anymore.
and Theology
Number of unique visitors:
Blogged by Clark Goble