You may be alarmed to hear that what you're about to learn about -- semiotics -- isn't really a "literary" theory at all.
"BUT WAIT," you might ask, "Isn't this supposed to be a guide to LITERARY theory???" Well, yes, and no. The thing is that a lot of schools of theory and criticism didn't originate in the field of literary studies, per se, although they are often related to it. Also, semiotics ultimately had a huge influence on subsequent literary theories in many ways, so that's why it's necessary to introduce it to you here.
Saussure was a pioneering linguist, so his "semiotics" (the science of "signs" -- more on these later) was used as a way to study systems of human language. However, you can apply semiotics to just about anything. The connection to literature seemed obvious, though, since literature wouldn't be much without language, now would it?
As mentioned in the introduction to this section, structuralism and semiotics were often used to turn non-scientific, subjective fields (as linguistics was once considered to be, for example) into something a bit more "science-y" or objective. So, to this end, Saussure's goal was to turn language into an object of scientific study. Until he came along to save it from its subjective ways, the study of language was too focused on the speech of individuals, according to Saussure. That is to say, specific instances of people using language. Saussure (pronounced sah-SOOR) argued that this would just be a nightmare for the field -- it's too wishy-washy, too subjective, and too subject to change over time.
So, how to make the study of systems of language more objective is the question. That should be easy, right? You just have to keep in mind that language is a structured system of signs dependent on social convention. You also have to study language synchronically instead of diachronically. ... [crickets chirp] ... What on earth does that mean? It's time for some definitions.
Synchronic vs. Diachronic
To study a system of language in a synchronic way is to look at the whole system as it exists at a single point in time. This would mean looking at every piece of the language puzzle -- the lexicon, grammar, syntax, etc. -- as a whole, as it exists simultaneously at one point in time (usually the present).
To do diachronic analysis is to look at how it has changed over time, historically. This is the realm of "historical linguistics." How has human language, or one specific language or dialect, changed over time? This can be fascinating stuff, but it's not what Saussure is worried about with his structural linguistics.
Now, all that seems pretty sensible. But as a final note, you will often see texts discussing two other associated terms from Saussure's work which would be useful to define here in case you encounter them in the future. Those terms are langue and parole.
It sounds fancy to use these French terms instead of their English translations, but basically they mean language and speech, respectively. Scholars keep using Saussure's original terms because they hold a specific meaning in the context of his theories, so direct translations are not always specific enough. This is not the first time you will encounter this in your study of theory, unfortunately, so just try to commit these things to memory.
When you talk about langue or language, you're doing a synchronic analysis. You're looking at the whole system of language as it is at one time. But talking about parole or speech requires diachronic analysis, because speech is time-sensitive. A quick way of saying this is that "Parole is the unfolding in time of a set of possibilities represented by langue." This makes sense because before you speak, you have to choose from a literally infinite number of possibilities for each part of your speech. You have to choose the words, the order to put them in, the tone with which you say them, which words not to say.. and so forth. Then when you say them, they're out there, in time, sometimes before you even know it. (That's why you get into trouble if you don't put enough thought into what you say before you open your mouth to speak... We've all been there before.)
All this is often summarized in something like the graph you see above. The horizontal axis (or "x" to the mathematically inclined) represents time, so studying language along this axis would be diachronic analysis, or looking at parole. Staying at "zero" on the x-axis or not considering the historical dimension is a synchronic analysis, or looking at langue.
SO -- synchronic, langue. Diachronic, parole. We've got that down -- now let's move on to the more fun stuff: SIGNS.
Signifiers and Signifieds
Someone once wrote me a letter (what? snail mail?!) that said the words written on the page were just "signifiers of signifiers." At first I thought -- is this person still writing in English? I had never heard this term before, and I thought for sure I must be missing something important here. In fact, to this day the word "signifieds" (as a plural noun) still comes up as a spelling error on my word processor.
That's because these terms have a very specific, technical meaning in the field of semiotics, and sometimes theorists just have to make words up to make their argument. This seems counterintuitive, but sometimes it makes more sense than you might think. So let's break these terms down a little bit, shall we?
It all comes down to signs, which are a huge part of what constitutes language. We're not talking about "no smoking" signs here, but two-sided entities that consist of a signifier and a signified. The signifier could be a word (or in technical terms the "sound image") which is supposed to represent the signified, which is the concept of the thing you're referring to. The signified is NOT an object, but it can be the concept of an object. So when you say a word out loud, you're signifying the signified with your signifier. Now that's significant!
To illustrate this, scholars often reproduce a version of a graph from Saussure's lectures which is helpful in visualizing this relationship. However, the standard example they almost always use is for the sign that represents [tree] -- you have to put it in brackets when you're talking about the sign, not just the word or sound (the signifier!) that represents it. In this example, though, I'm going to use the sign [fox] because it's just, well, foxier.
The arrows are to show that the two levels are two sides of the same coin -- or sign, in this case. Because each sign is made up of a signifier and a signified. Okay. The top level is the signifier -- the sound image -- which could be the sound of someone saying "fox," the word "fox" written on a page, or even an image of a fox (which tend to be pretty cute, I have to say). The bottom level is the signified, which is the concept in your mind of "fox." Not some specific fox that you're seeing in front of you, but the idea of one. Got it? Good! Because it gets a little bit harder after this.
It wasn't a new idea by Saussure's time that words or images referred to concepts instead of objects out in the real world. But there are two important things that Saussure has to say about the sign:
There is an arbitrary relationship between the signifier and the signified, as well as the sign and the real-world referent.
We only know what one sign is because it's different from all other signs.
Now, "arbitrary relationship" might sound kind of scary at first, but it actually makes a good deal of sense. All it means is that there is no reason why the word/signifier "fox" should represent the concept of fox, so it's perfectly sensible for the French to say "renard," the Finnish to say "kettu," or the Irish to say "sionnach." There's also an arbitrary relationship between the spoken word and the way it appears as written language (although some languages began with pictorial representations of words that resembled their referents).
If it weren't for this "arbitrary relationship," there would only be one language spoken on the planet, because everyone would just say the one "natural" non-arbitrary way to say things. There is just no such thing as a "natural" sign! These things only carry the meaning they do because everyone in a language community agrees that they do -- it is simply a matter of convention.
The second point is even more important, because it is an important aspect of structuralist theories. You understand the meaning of a sign "negatively," never "positively." In other words, you don't know what [fox] means because it inherently holds that meaning. You only understand it because this sign is different from all other signs. For example, the signifier "fox" is different from "fax," "fix," "for," "box," and so forth. The signified fox is also different from wolf, dog, cat, and so forth.
Alright! So now we finally can get to the bottom of what my pen-pal meant when she was talking about signifiers of signifiers. The words themselves are signifiers of the concepts/signifieds, and the written words are signifiers of the spoken words -- so, that's two levels removed from the original idea or meaning behind what's written on the page. We can also now see the importance of knowing and using specific terminology when talking about such complex ideas: it just takes too long to say these things otherwise! Theorists aren't trying to confuse you by writing this way (I think). They're just trying to save time.
So is your mind blown yet? Next, we're moving on to Roman Jakobson, a Russian theorist who was heavily influenced by the work of our now dear friend Saussure.
Review Your Terms:
semiotics synchronic diachronic parole langue
sign signifier signified arbitrary relationship
References
Eagleton, Terry. "Chapter 3, Structuralism and Semiotics." Literary Theory: An Introduction. Minneapolis: University of Minnesota Press, 2008. 79-109. Print.
(de) Saussure, Ferdinand. "From Course in General Linguistics." Trans. Wade Baskin. Norton Anthology of Theory & Criticism (1st Edition). Eds. Vincent B. Leitch, et. al. New York: W.W. Norton & Co, 2001. 960-977. Print.