What is Linguistics?
linguistics is a rigorous discipline with established methods which strives
to follow the scientific method. Our success at being "scientific" is variable
and to a large extent a matter for debate, but it is important that competing
theories and approaches can be compared, at least in principle, on the basis of
how well they satisfy criteria of scientific rigor. |
Second, the primary object of linguistic study is human language, not language in other extended senses to be discussed below.
Third, it is the study of language not languages . While we are concerned with the structure and properties of specific languages, our ultimate goal is to understand the properties of language in general. For example, we are less concerned with the quirks of English nouns or Greek verbs work than with how these compare to things in other languages and what they can tell us about nouns and verbs universally. A lot of the time it is only possible to understand why a specific language does some apparently odd thing by considering general properties of language that can only be observed in other specific languages.
Fourth, as is implied by the vague formulation of the definition, linguistics is an extremely broad discipline with an array of subfields. A good way to get an idea of what linguistics really is is to take a look at what these subfields are and how they fit together.
The first sound in English tall and the first sound in Spanish tu 'you' are similar in several respects, but they differ in that the English sound can be described as alveolar (being pronounced at the ridge behind the teeth) and aspirated (being accompanied by a puff of breath which you can feel if you hold your hand in front of your mouth when you pronounce it), while the Spanish sound is dental (being pronounced at the teeth) and unaspirated (without the puff of breath).
In English, the sounds we represent as /p,t,k/ are aspirated (with the puff of breath) at the beginning of a word, as in pill, tall, kill , but not when they come after an /s/, as in spill, stall, skill . You can test this by pronouncing the pairs with your hand in front of your mouth. The difference in pronunciation is a phonetic fact, but the rule describing it is a phonological rule that describes the English sound system. There are plenty of languages that do not have this rule.
There are two meanings for the word unlockable depending on its structure:
Note that the following sentence is actually ambiguous, depending on how we interpret the relationship between ten and didn't :
Ten delegates didn't attend the session.
These sentences can all express the same request, but often indirectly.
Human language vs. other meanings of language
Like most other words, language has a number of meanings that are somehow related but turn out to refer to very different things. Note for example the range of meanings in this dictionary entry:
For most linguists, language is the pattern of human speech, and -- what most distinguishes linguistics from the everyday use -- the (implicit) systems that speaking and listening rely on.
Other phenomena come to be called "language" because of more or less close connections or analogies to this central case: writing, sign languages, computer languages, the language of dolphins or bees. The ordinary-language meaning of the word reflects this process of extension from a speech-related core.
Linguistics is concerned with language in the latter sense only to the extent that things like computer languages can help us to understand language in its core sense of normal human speech. An exception to this is sign language which, aside from being realized in the form of visual gestures instead of auditory gestures, are in every respect full-fledged human languages, and thus are a central, rather than peripheral, object of linguistic study.
The central role of spoken language, as seen by the first definition in the dictionary entry, is also reflected by the source of the equivalent word in various languages. One common strategy is simply to use the word tongue to express the concept "language": it's one of that organ's most obvious functions in modern humans.
The core of the field of linguistics has always been the analysis of spoken linguistic structure, that is, we are more interested in language as it is normally spoken than in language as it is written or is "supposed" to be written, on which more below.
In common usage, people usually use "dialect" to mean a nonstandard variety of a language. This can be seen in the first definition below, which uses "language" for the standard variety. (Cf. definition 1c for "language" above.)
So non-linguists often make this distinction:
For linguists, however, no one actually speaks an entire "language"; rather, every utterance occurs in some (social and geographical) dialect.
For convenience, we can define a language as a group of mutually intelligible dialects. That is, while the dialects differ in detail, the speakers can all understand each other. But this is only a convenient abstraction, not a technical concept to which linguists attach any special importance, because when we sit down and actually try to draw lines between languages we run into serious problems:
In fact, the classification of dialects into languages that we are all familiar with is based at least as much on political criteria as on linguistic ones. The most famous formulation of this is due to Max Weinreich:
A language is a dialect with an army and a navy.
The crucial point for us is that, as far as linguists are concerned, all dialects are equal. Nonstandard dialects may lack prestige, but not logic or communicative function.
The central thesis of Pinker's book is that language is a type of instinctual behavior, not an acquired skill. The forms of language spoken by all humans (setting aside brain traumas and the like) are of equal status as examples of the innate ability to learn and use language.
Remember, for linguists the term "dialect" refers to a specific variety of language so everyone speaks a dialect, whether it's considered a prestigious one or not.
For example, informally I can say that I speak English.
But what I really speak is a particular variety of English, what one might call General American, but even that's an abstraction.
When you really get down to it, everyone actually speaks an idiolect: the specific form of speech of one individual. "Dialect" is a lower-level abstraction over similar idiolects, and "language" is a higher-level abstraction.
Fortunately, though, since language has to be usable for communication, the idiolects of people within a given community are close enough to each other that, for many purposes, we can treat them as the same. But at times it will be important to remember the differences
There are plenty of valid controversies about language.
Some questions are entirely political: should governments try to accommodate speakers of minority languages? how important is it to maintain rigorous standards of usage? is it bad to borrow words from another languages rather than inventing native ones?
Other questions are factual, though they have immediate practical consequences: does bilingual education work? what are the consequences of oral education for deaf children? to what extent can ordinary citizens understand legal contracts? how well do computer speech recognition systems work?
set of questions are mainly interesting to those who care about language itself:
Reasonable and informed people can and do disagree about
these and innumerable other linguistic issues. Particular arguments may be illogical,
or particular claims may be false, as in any debate, but our state of knowledge
leaves room for a range of opinions.
In some cases, the "flat earth" position is only held by exceptionally ignorant people, and gives rise to jokes with punch lines like "if English was good enough for Jesus, it's good enough for me." However, there are plenty of misconceptions about language among otherwise reasonable people, not to mention French literary theorists or popular pundits. These are worth calling myths.
Here are some examples of linguistic myths which you can check out at the sci.lang FAQ:
A couple of others we've already touched on above but deserve to be repeated:
Myth: speech and writing are parallel forms of linguistic expression,
different but equally fundamental types of text.
non-standard dialects are degraded and errorful versions of standard languages.
Myth: Primitive cultures have primitive languages, at
a lower level of development and less well able to express a wide range of ideas.
As mentioned above, the fundamental point of the text for this class, Steven Pinker's The Language Instinct , is that human linguistic ability is a kind of instinct (though far more complex than the phenomena this word often evokes).
In chapter 2, he gives several kinds of evidence for this position which we'll look at below.
Universality and equality
We have no record of a human culture that lacked language. Furthermore, all languages that we have been able to examine seem to be of roughly the same complexity and to have roughly the same expressive power. This is quite different from familiar aspects of culture.
There are cultures which are objectively primitive technologically. Material innovations are constantly undertaken and passed from one culture to another and have been as far back as we can go, from the use of specific metals for making tools, to the adoption of sedentary agricultural methods, to the use of computers. This is not to say that certain cultures are better than others or to make any sort of value judgment, it is rather the objective observation that culture A has some specific skill or knowledge wich culture B does not. Culture A may be more advanced in certain areas while culture B is more advanced in others.
But there are no primitive languages.
Different languages do have different amounts of complexity in different areas, but things balance out overall. There is no correlation between the complexity or "logic" of a language and the material sophistication of the people who speak it.
For example, the familiar languages of Europe have a single word corresponding to English we which means "me and at least one other person", but many languages, spoken by people from considerably more primitive material cultures have several words which allow their speakers to be more clear on who else is included with "me". For example, Mohawk, a Native American language, has the following prefix forms:
In certain situations of language contact, a pidgin may arise to serve as a lingua franca or means of communication among people who don't share a common language. A pidgin is not a real language: it has no native speakers, and its grammar is quite simple.
An example is an earlier form of Tok Pisin ("talk pidgin") which arose in Papua New Guinea: a simplified form of English influenced by native languages of the island. This quote is from the 19th century.
Sometimes a pidgin can come to serve as the first language of children born into the contact situation. What happens in this case is creolization: the pidgin develops rapidly from the simple pidgin to a much more complex creole, a true language. As Pinker discusses, it seems that the children are imposing the innate linguistic ability on the impoverished input and coming up with a language worthy of their lingustic abilities.
Creole: A creole language, or just creole, is a well-defined and stable language that originated from a non-trivial combination of two or more languages, typically with many distinctive features that are not inherited from either parent. All creole languages evolved from pidgins, usually those that have become the native language of some community.
This process can be observed in the development of Tok Pisin. In the course of the 20th century, the language has developed far greater complexity in a number of ways. One is that verbs take a subject marking element i (from English "he") even in the presence of a full noun phrase. This is unlike English, but very much like many other languages, including Spanish, German and Choctaw.
Similarly, verbs take a suffix im when they are transitive (i.e. have a direct object). It's historically from English him , but it serves a very different function now, being there in addition to the object:
It's much more like transitivizing suffixes found for example in a number of native American languages like Nahuatl (the language of the Aztec empire).
The essential point is that this grammar has developed independently of the English on which it is largely based. Scholars debate the degree to which the new grammar derives from neighboring languages or from innate human grammar, but either way it shows the effect of the instinct for grammatical language.
Contrary to popular belief, children do not seem to learn language the way they learn to tie their shoes or draw pictures. They're way too good at it. They deduce a complicated grammatical system on the basis of a very small amount of noisy data, in a way that looks suspiciously like they already know certain things basic properties of human language from the beginning.
It's not difficult to find examples of children creating new words that they've surely never heard before, yet based on the principles of the language around them. Overgeneralizing rules for word formation is a common example:
This ability can actually be tested experimentally. A child can be presented with a task requiring the creation of a new form of a word never heard before -- the classic example involves showing children a small bird-like creature with the made-up name "wug", leading to the term wug test.
People with Specific Language Impairment (SLI) - a disorder characterized by a drastic disruption of linguistic ability without corresponding disruption of general cognitive functions - have great difficulty with this seemingly simple task.
There is evidence from language disorders that language is consistently located in specific parts of the brain, and perhaps in some sense these areas are designated for language. That is, damage to specific physical areas in the brain correlates with specific impairments in linguistic ability.
One syndrome called Wernicke's aphasia typically results in fluent but nonsensical speech. Affected individuals often sound superficially to be speaking normally, but upon examination what they produce, while frequently containing strings that conform to the grammatical rules of the language, has no coherent meaning, and typically includes made-up words. They also have trouble understanding the normal speech of others.
This aphasia is the result of damage to a region of the brain called Wernicke's area, near the primary auditory cortex that controls hearing.
The fact that damage to Wernicke's area leads to similar deficits in different people, speaking whatever their native language might be, already suggests that a specific area of the brain may have evolved to deal with language processing.
Quite interestingly, however, the linguistic function of this area does not seem to be tied to sound in the modern brain, as shown by the evidence of deaf aphasics.
For users of sign language, damage to Wernicke's area leads to fluent but nonsensical signing, as in the following translated example.
By contrast, a signer with damage to Broca's area has trouble articulating signs and cannot communicate fluently -- just as in spoken language it leads to stilted, labored speech (Pinker, pp. 34-35).
Cases such as this provide striking evidence that linguistic abilities are part of the brain's basic endowment, which when functioning correctly enables people to manipulate symbols in a structured way to convey meaning. It doesn't actually matter whether the language is spoken or signed. They also provide evidence that sign language is, in some important sense, the same as spoken language, a point which we will discuss in some detail in a later lecture.
Linguists talk a lot about "grammar", and this class will be no exception, so to conclude this lecture I want to say something about what linguists mean by this term and, more importantly, what they don't mean.
When you hear the word grammar it may conjure up bad memories of English class where you had to memorize a series of weird rules to make sure that you didn't make any mistakes while writing papers. This is what we call prescriptive grammar , because it's a matter of someone telling you what (and what not) to do.
This is not what linguists do.
Instead, we're concerned with what we call descriptive grammar
Now, there is a type of correctness that linguists actually are concerned with, a notion that is called grammaticality or acceptability .
The things that English teachers get mad about, or that mark certain dialects as "substandard" are usually quite arbitrary. In fact, the constructions they find bad are usually found in very similar form in other languages, where they don't raise an eyebrow. The reasons why some linguistic feature is looked on as "improper" or associated with low prestige typically have quite a bit to do with historical accident or whim, and rarely have anything to do with real linguistic issues.
The determination of what is grammatical or ungrammatical, on the other hand, is based strictly on scientific observation.
Three examples of prescriptivist silliness
As with everything, the issues here can be made more clear with some examples
A lot of what prescriptivists get uptight about, including the last example above, has to do with one of the central facts about language:
All living human languages are constantly changing.
In a later lecture we'll talk in detail about the how and why of language change, but for the time being, all that matters is that it is inevitable. Your children will not speak exactly like you do, and your ancestors in three hundred years or so, if they met you, would probably have a lot of trouble understanding you, even if your family stayed in the same place during that entire time.
In a number of cultures, people have gotten so concerned about the "decay" of their language that they've set up commitees or academies to counteract it.
So are we doomed to slide ever further down the slippery slope of language decay until we are reduced to an emotive jibber-jabber of grunts and squeals?
In spite of the fears and dire predictions of generations of prescriptivists and even some early linguistics, language change is not decay.
Just as all languages are equal in complexity and sophistication, so are different historical incarnations of a single language.
As far as we can tell, language is getting neither worse nor better, just different.
Now, good clear writing style and solid rhetorical skills are rather different matters. They are learned separately from language, although they use language, and they can very well get better or worse from generation to generation, and vary quite a bit from person to person.
But that is a story for a different course.