Sociolinguistics I:
how language and society interrelate through the observation of the use of language in everyday life.

Introduction.

Looking at Part II of Crystal's Encyclopedia, on "Language and Identity", we can see that he has taken an approach that starts from the biologically given (Unit 6, on "Physical Identity", and makes its way through "Psychological Identity" (Unit 7), and via "Geographical Identity" (Unit 8), to the broader social and cultural themes of national, ethnic, and social identity in Units 9 and 10 (we will deal with 11 and 12 later in the course).

This is a good approach for a book that places this set of topics almost at the beginning, when readers have not yet been exposed to the basic facts about the structure of phonology, morphology and syntax that we have introduced you to over the past three weeks. However, since you already have some knowledge of the way linguists approach language structure, you are in a position to take a more technical view of the interaction between language and society. In other words, you don't need much linguistics to understand most of what Crystal presents in Units 6 through 10; and he has done a brilliant job of presenting a lot of fascinating information to the lay public. What we will do in this lecture and the next is introduce you to another way of finding out how language relates to social life, a way that does require linguistics.

Since the early 1970's, the sociolinguistics program at Penn has developed a tradition of sociolinguistics that is an integral part of linguistics (rather than being, say, a branch of sociology or anthropology), and this approach is what many people now understand sociolinguistics to be.

Let's start with the question of what kind of information you get when you talk on the telephone to a stranger -- someone you have never met. In addition to the explicit information that is the content of the messages you receive during such a conversation, you get another kind of information. Usually can tell the speaker's gender, and often you can make a pretty good guess as to his or her approximate age. Other things you may be able to guess are whether or not the person is a native speaker of your language, what part of the country she comes from, whether or not he is highly educated, and possibly her ethnic or racial background, or his social class. In addition, you may get some idea of her attitude to you (personal, businesslike, formal, friendly, etc.), and of his feelings about the topic of conversation (competence, level of interest, and so on).

Your ability to make such inferences, sight unseen, about speakers and their intentions, depends only in small part on biological givens. Most of it is shared knowledge that you have acquired in the process of acquiring language, as a member of your speech community. (Imagine yourself on the telephone with a speaker of a language you don't know -- let's say Turkish -- and you'll quickly discover that you would probably be limited to making inferences about age and gender of your interlocutor.)

The fact that you can make many correct social attributions about people on the basis of their speech has to mean that speech communities are internally differentiated. After all, if everyone in the community spoke exactly the same way, no such subgroupings would be identifiable, nor would we be able to correctly interpret speakers' intended politeness or impoliteness, formality or informality, etc. We distinguish two broad axes of variation in the internal structuring of speech communities: the inter-individual (that differentiates groups of people) and the intra-individual (the same person's differential use of language in different circumstances). Even in small village communities with minimal social differentiation, where people practice a traditional way of life, sociolinguists have so far not found any speech community that is internally undifferentiated, nor any mono-stylistic speakers.

The question is: how does this work? What are the characteristics of a person's speech that carry this social information? Can the same linguistic features that differentiate along the inter-individual dimension also be used to differentiate along the intra-individual dimension? And most importantly, how can we find out?

We will approach these questions via a close case study of one feature of English that is involved in both the inter-individual and the intra-individual axes of differentiation of the speech community, looking first at how two well-known authors have represented social differences via speech. But before launching into that case, we'll need a few basic concepts of sociolinguistics.

vernacular: Crystal tells you it's "the indigenous language or dialect or a community". We'll make it a little more precise by thinking of it (following Labov) as:

* the earliest learned, and most basic, linguistic system acquired by an individual during childhood language acquisition. It's your "native language", the one you can most easily and unreflectingly access without the meta-level instruction in literacy or language arts that you learn later in life. Since you learn it in childhood, it reflects the language of the community you grew up in, which has a geographical location.

As you go through life, you continue acquiring language. You may learn other languages, but you also learn more about your native language, including a layer of conventions of the written language that may be relatively close or relatively distant from your vernacular. Every speaker thereby ends up with a linguistic repertoire that includes the entire inventory of their linguistic knowledge, along with the knowledge of how to deploy these different linguistic resources differently according to the social situation. (Just think of the subtle ways you may alter your language when talking to a best friend, a grandmother, someone giving you religious instruction). John Gumperz described linguistic repertoires as being the property of both communities and individuals, and as having a dialectal component (maybe not the best choice of words, but meaning the locally-situated vernacular), and the superposed component (what you learn later in life).

There are a few more concepts that will come in handy, but we'll deal with them as we go along.

A CASE STUDY

Part I. Literary sources.

To see the complete versions of the novels excerpted here, you can visit the Public Domain HTI Modern English Collection at the University of Michigan.

First, let's examine two excerpts from Mark Twain's Huckleberry Finn, first published in 1885.

In the first excerpt, Huck's father, "Pap", a no-account, abusive ne'er-do-well who has never taken any interest in Huck, has emerged from oblivion to claim the money he has heard that Huck has come into:

Huck Finn Excerpt 1.

1. "Looky here--mind how you talk to me;

2. I'm astanding about all I can stand now

3. --so don't gimme no sass.

4. I've been in town two days,

5. and I hain't heard nothing but about you bein' rich.

6. I heard about it away down the river, too.

7. That's why I come.

8. You git me that money to-morrow

9. --I want it."

In this first excerpt, Twain has represented many dialect features to portray the roughness of Pap's character. These include:

An author who was a keen observer of Missouri dialects, Twain gives us plenty to look at, but in this case study we are going to focus on only one feature, the one popularly known as "dropping the g". In Excerpt 1, Pap "drops a g" in bein', but retains it in astanding.

Here is a second short excerpt from later in the novel, when Pap has imprisoned Huck, and wakens from a drunken stupor to find Huck with his gun.

Huck Finn Excerpt 2.

1. Pap: "What you doin' with this gun?"

2. I judged he didn't know nothing

3. about what he had been doing, so I says:

4. (Huck) "Somebody tried to get in,

so I was laying for him."

Although Huck uses a double negative in his first-person narration, he is much closer to the standard in his choice of didn't, rather than hain't, as a past tense auxiliary. In addition, we notice that although Pap once again "drops a g", in doin', Huck does not, despite his choice of the colloquial expression laying for to mean something like "getting ready to retaliate".

Before trying to analyze what is going on with this "g-dropping", we'll look at the work of another English author, the British D.H. Lawrence, in Lady Chatterly's Lover (1928). In this excerpt, Lady Chatterly (Connie), first encounters her husband Clifford's gamekeeper, Mellors.

Lady Chatterly Excerpt 1.

In this excerpt, Lawrence actually tells us when Mellors is switching into a different part of his repertoire. Though he has used -ing in his 'Good morning, Sir', to Lord Clifford, he "drops into" the broad sound of the vernacular as he bids good morning to Lady Connie. This pattern is continued in Excerpt 2, when Connie and Mellors meet next.

Lady Chatterly Excerpt 2.

Mellors is apparently behaving somewhat disingenuously here, as he seems perfectly capable of speaking "ordinary English" when he is discussing his work, whether with Lord Clifford or with Connie, as we see in Excerpt 3.

Lady Chatterly Excerpt 3

Finally, we see in the conversation between Connie and Mellors after they have made love, that Mellors is consistently represented as using the "vernacular", including -in rather than -ing, in his intimate relations with Connie.

Lady Chatterly Excerpt 4

Summary of -in vs. -ing in Huckleberry Finn and Lady Chatterly.

In a sample of about half the chapters of Huckleberry Finn that I examined, Huck does not once use -in, holding steadfastly to -ing. Both Pap and Jim, Huck's fugitive slave companion, are portrayed as varying in their use, although both use more -ing than -in in the -ing words.

With Lawrence, we see that the nobility (Clifford and Constance) never "drop a g", whereas Mellors varies strictly according to whether he is speaking "proper English" or "the broad vernacular" of Midlands England: 100% -ing in "proper English", and 100% -in in the "vernacular" (which Lawrence usually underscores by mentioning the switches explicitly). A fourth character, the village woman Mrs. Bolton, is like Pap and Jim in occasionally "dropping her g's".

Apparently these authors are using their readers' presumed deep knowledge of the social meaning of the -in vs. -ing alternation in English in their characterizations. If we are to go by other things we know about the characters, the alternation seems to represent (at least): social class, education, race, and intimacy.

Part 2. The study of the speech community.

We are now ready to approach what is going on from a linguistic point of view. A first question is:

What kind of alternation is this, in terms of the structure of the English language?

A. "Dropping g's"? Maybe the folk label for what is going on is essentially correct. Are people "dropping the g" when they say goin'? Certainly its representation "drops off" in writing! But in the spoken language, we normally say /ng/(this is a convenient typographic notation for a single segment, the velar nasal which is written in phonetics as an "n" with the curly tail of a "g" on the right: you can see it in the IPA chart in the Phonetics/Phonology lecture notes), not /g/. Pronouncing going as /goingg / (sometimes represented in writing as "goink") is nonstandard and quite marginal within the English-speaking world. Actually, the alternation in pronunciation is between the apical nasal /n/ and the velar nasal /ng/.

B. Phonology?One hypothesis might be that there is an overall alternation between /n/ and /ng/ in English. We might think this is the case, for example, by observing that the negative prefix in- sounds like alveolar /n/ when it is followed by an alveolar /t/ or /d/ (intractable), but like velar /ng/ when it is followed by a velar /k/ as in incredible. But this is not a tenable hypothesis. Ran /ran/ and rang /rang/ are different words, as are sin /sin/ and sing /sing/. Theinescapable conclusion is that /n/ and /ng/ are separate phonemes and do not alternate freely.

So where can they alternate? Let's try to narrow it down. It is easy to narrow it down to word-final position (you can't say /sining/ for singing, whereas you can say /singin/. It's also easy to establish that it never occurs after vowels other than /i-/. So, maybe it's the -ing suffix.

C. The -ing suffix? Many of the words where we noticed authors writing -in' are words where -ing is a suffix: participles like doing and waiting. But the suffixal status of the -ing in morning is a little less clear, and nothing proves that the suffixal analysis is just wrong, since the morphemes of nothing are 'no' and 'thing', not *'noth' and -ing.

D. You guessed it! in class. The -ing whose pronunciation varies between /n/ and /ng/ in every dialect of English is the "unstressed -ing" that occurs at the end of every present participle, every gerund, and many other kinds of words: prepositions like during, common nouns like ceiling, proper nouns like Reading -- literally thousands of English words that crop up with great frequency in everyday conversation.

Unstressed -ing in the Speech Community