1.1.1. Language as a scientific subject matter.
Language is everywhere around us. The ability to use language is a basic skill that we are all endowed with to live as human beings. It is hard to imagine how one could live without some form of language. But we tend to take things around us for granted, as if there is nothing particularly remarkable about our species-specific ability to use language. In our study of linguistics, we try not to take things for granted. In linguistics, we try to make careful observations about language, pose questions about those observations, make hypotheses to explain them, test the validity of those hypotheses, and revise, confirm or extend those hypotheses. Such activities of research are typical of the natural sciences.
--metal color, Max Planck, quantum physics
--apple dropping, gravity
--other things taken for granted
Indeed, we take linguistics to be a science, like any of the natural sciences. Although the subject matter of linguistics does not concern natural phenomena (which are the subject matters of the "hard sciences"), but deals with humanistic and social phenomena, the approach we take is no different from that of a natural science.
What goal do we have in the study of language as a scientific subject matter? Noam Chomsky, the architect of modern linguistics, says that language is a mirror of the mind. In this course, we use language as a probe into the nature of the human mind. By studying the nature of human language, we gain valuable insights into one important aspect of human cognition.
1.1.2. Object of inquiry about knowledge of language.
a. Content, acquisition, and use. What are the things we observe about language that we try to explain? There are at least three kinds of linguistic phenomena that are remarkable enough to call for an explanation. First, we possess a remarkable amount of knowledge of our language(s), which allows us to communicate with others in a theoretically unlimited way: that is, we can understand utterances that we have never heard before, say sentences we never used before, etc. (as will be illustrated in more detail below). Although our experiences seem vastly different, the knowledge we have about our language clearly must be sufficiently the same among us to allow for successful communication to take place. Finding out what the content of that knowledge is and characterizing it in the most appropriate way is the primary concern of (core) theoretical linguistics. Secondly, we as human beings seem to acquire the knowledge about our language(s) with remarkable speed and efficiency. Typically, children gain control of most of the essential aspects of their language in 2-3 years by the age of 5-6. This is extremely remarkable especially given the highly complex nature of linguistic knowledge (as we shall see in the rest of this course) and the limited experience the child has been exposed to. [......] Thirdly, we can distinguish between the content of our linguistic knowledge and its use (between our linguistic competence and linguistic performance.) Our linguistic knowledge (competence) is theoretically unlimited, but our use of it (comprehension and production) is limited.
b. Tacit knowledge and formal knowledge. Most of our linguistic knowledge is tacit, and it's the kind we tend to take for granted. As a linguist, we try to characterize that tacit knowledge in a formal way, so that our unconscious knowledge of a layman becomes the conscious knowledge of an educated specialist.
c. Components of grammar. Throughout this course, our concern will be with the content of our linguistic knowledge. When we hear people say that a person knows a language, we shall say that he/she possesses the grammar of his/her language. For the convenience of description, and justified on extensive empirical grounds, a grammar can be divided into several components, or modules. The elementary divisions of a grammar are a syntax, a semantics, and a phonology. [......] There is also a component of morphology, dealing with the structure of words.
1.2. Knowledge of ÝLanguage
What does it mean to say that you know a language? Let's start out by looking at some examples that illustrate our knowledge. In this section we shall illustrate a speaker's knowledge of the phonology, morphology, syntax, and semantics of his/her language:
1.2.1. Phonological Knowledge
The following examples illustrate some aspects of your knowledge of English phonology:
(1) George Bernard Shaw and the English spelling system: how do you pronounce ghoti ?
(2) English phonotactics: brick, blick, bnick, sdpick.
These examples show that you know how a word can be pronounced even if you have never heard it said before, and whether a given sound sequence is a possible sequence in English even if you have never encountered the sequence before.
1.2.2. Morphological Knowledge
Your knowledge of English morphology is amply illustrated by Lewis Carroll:
(3) Lewis Carroll and Through the Looking Glass (1872) and Alice in Wonderland (1865).........................
Twas brillig, and the slithy toves
Did gyre and gimble in the wabe:
All mimsy were the borogroves,
And the mome raths outgrabe.
Although some of these words are non-occurring English words, you know that they are all potentially meaningful and legitimate words of English. In fact, you also know what parts of speech (category) some of these words belong to, and may even imagine what they might mean.
1.2.3. Semantic Knowledge.
a. Ambiguity. Your knowledge of English allows you to discern the ambiguity (or lack thereof) of some of the sentences below:
(4) a. Flying planes can be dangerous.
b. Flying planes are dangerous.
c. Flying planes is dangerous.
(5) a. Visiting relatives can be a nuisance.
b. Visiting relatives is a nuisance.
c. Visiting relatives are a nuisance.
(6) The chicken is ready to eat.
(7) Mistrust wounds.
b. Synonymy. Or synonymy among certain others:
(8) a. John has seen Bill.
b. Bill has been seen by John.
(9) a. It was certain that Kristel would win the Best Soprano Award.
b. Kristel was certain to win the Best Soprano Award.
(cf. Kristel was certain that she would win the Best Soprano Award.)
c. Missing Information. It also allows you to fill in missing information in a systematic (i.e., constrained and not random) way:
(10) a. John is eager to please.
b. John is easy to please.
(11) a. Visiting relatives can be a nuisance to him.
b. Visiting relatives is a nuisance to him. (him = visitor)
c. Visiting relatives are a nuisance to him. (him =\= visitor)
1.2.4. Syntactic Knowledge
a. Word Order. As a speaker of English, you know how the parts of a sentence should be arranged. For example, you know that, of the 6 logical possibilities in (12), only 2 are allowed in English:
(12) a. John saw Bill. (SVO)
b. *John Bill saw. (SOV)
c. *Saw Bill John. (VOS)
d. *Saw John Bill. (VSO)
e. *Bill saw John. (OVS)
f. Bill, John saw. (OSV) (with comma after Bill)
If you speak Japanese or Navajo, however, then you know that the orders corresponding to (12b), (12f), are possible, but the others are not. This is illustrated by the Japanese examples below:
(13) a. *John-ga mit-a Bill-o. (SVO)
John-subj see-past Bill-obj
b. John-ga Bill-o mit-a. (SOV)
c. *mit-a John-ga Bill-o. (VSO)
d. *mit-a Bill-o John-ga. (VOS)
e. *Bill-o mit-a John-ga. (OVS)
f. Bill-o John-ga mit-a. (OSV)
Chamorro, Irish and many other languages differ from English and Japanese, etc., in allowing the orders represented by (12a) (SVO order) and (12d) (VSO), and excluding some or all the others. Latin, however, allows all the six logical possibilities:
(14) a. Ami1cus Caecilium salu1tat.
friend Caecilium greets
'The friend greets Caecilius.'
b. Caecilium ami1cus salu1tat.
c. Ami1cus salu1tat Caecilium.
d. Caecilium salu1tat ami1cus.
e. salu1tat ami1cus Caecilium.
f. salu1tat Caecilium ami1cus.
The constraint on word order that is part of your knowledge of English is further illustrated by the examples below:
(15) a. a gray-haired student of physics
b. a physics student with gray hair
c. a gray-haired physics student
d. a student of physics with gray hair
e. *a physics gray-haired student
f. *a student with gray hair of physics
While English is fairly liberal in allowing 4 out of the 6 possibilities in (15), Japanese and Navajo allow only the form corresponding to (15c), rejecting all other word orders.
b. Grammaticality. Your knowledge about whether a given string is a grammatical utterance in your language is not limited to that related to word order, but is manifested in many other ways. For example, regardless of the fact that it "does not make sense", you know that (16a) is grammatical, but (16b)-(16c) are not:
(16) a. Colorless green ideas sleep furiously.
b. *Sleep colorless furiously ideas green.
c. *Furiously sleep ideas green colorless.
You can make fairly subtle judgments concerning the grammaticality of sentences, regardless of whether they have ever been heard before:
(17) a. ??What book did you wonder why John bought?
b. *Who did you wonder why bought the books?
c. ??This is the book which I wonder why John bought.
d. *This is the person who I wonder why bought the books.
c. Structure and Categories. More importantly, in the area of syntax--the subject matter of this course, anyone having competence in a language can have intuitions about the structure and categories of grammatical sentences of his/her language. They can identify the words in an utterance, but they also can see the sentence as more than merely a linear arrangement of words. Observe the following sentences.
(18) a. John slept.
b. John left home.
c. The boy left home.
d. Colorless green ideas sleep furiously.
e. The man who was mixing it fell into the cement he was mixing.
f. The quick fox jumps over the brown lazy dog.
g. The old men and women left early yesterday.
h. John talked to the man in blue and the woman in red.
Regardless of the number of words within a grammatical sentence, speakers of English generally have no problem dividing it into two "natural" sub-parts, or immediate constituents (ICs). This is trivially the case with (18a). But speakers also have no problem cutting the sentences (b-h) into two ICs. For example:
Indeed, they can analyze a given sentence into successive layers of ICs, regardless of whether a given grammatical sentence makes sense or not:
These IC charts provide a visual representation of the speakers' intuition that certain subparts of a sentence form natural groups to the exclusion of others. For example, in (20), the and boy form a constituent to the exclusion of the other words, as do left and home. There is a closer relationship between the and boy than between boy and left. Similarly, (21) captures the intuition that green is more closely related to ideas, than to colorless, than to sleep, etc. These hierarchical relationships among elements are not directly represented by the linear strings in (18), but they are quite naturally expressed by these IC charts. An notationally equivalent way to represent these intuitions is to use a branching tree diagram. Thus (21) can be converted to the equivalent (23) below:
In addition to these intuitions about the constituency of elements within a sentence, speakers also know that the constituents of each string belong to one category (part of speech) or another. For example, the constituent colorless green ideas seems to belong to the same category as the constituents John in (19), the boy in (20), and the man who was mixing it and the cement he was mixing in (22): they are all Noun Phrases (NP). On the other hand, sleep furiously belongs to the same category as left home, fell into the cement he was mixing, left early yesterday, jumps over the brown lazy dog, etc.: they are Verb Phrases (VP). Furthermore, colorless, green, lazy, etc. belong to the category of an adjective (A), sleeps, fell, mixing, left are verbs (V), and ideas, home, boy, man are nouns (N). (Some of you may not know what to call these constituents--these are technical terms given by grammarians--but you will agree that ideas and home belong to the same category, but not, say, ideas and fell.) We can encode the intuitions about both structure and category in a labeled branching tree diagram:
(We shall have more to say about these tree diagrams later, but will ignore the details now.) Similarly, the sentence (18e) can have the following representation:
Another way to represent speakers' intuition about structures and categories is by means of labeled bracketing. The tree in (24) can be converted to the set of labeled brackets in (26), and the tree in (25) to the set in (27):
(26) [S [NP [A colorless] [NP [A green] [N ideas]]] [VP [V sleep] [Adv furiously]]].
(27) [S [NP [NP [Det the][N man]][S [NP who][VP was mixing it]]] [VP [V fell]
[PP [P into][NP [NP [Det the][N cement]][S he was mixing]]]]].
In this chapter, we have seen that speakers of English have remarkable knowledge about their language, some of which we have not been aware of but all of which are clearly quite systematic and consistent among speakers. In particular, you know what is and what is not a possible string of sounds in English, what is and what is not a possible English word, or a possible English sentence, and you are able to detect various semantic properties among grammatical sentences (synonymy, ambiguity, missing information, etc.) Furthermore, in spite of the linear arrangement of elements within a sentence, you know that the elements of a grammatical sentence are hierarchically structured, in that certain sub-strings of a sentence form natural units to the exclusion of other elements. Sentences are seen to consist of immediate constituents each of which may consist of its own immediate constituents, etc. This intuition that you have is evidenced by the fact that you can systematically break up a grammatical string into successive layers of immediate sub-strings.
We have also seen that our intuitions about structure and category may be captured visually by means of IC charts, tree diagrams, or labeled brackets. These devices provide a means to represent the tacit knowledge that speakers have about the syntactic structure of their language(s). Having a conscious knowledge of what you already tacitly know, and being able to talk about what that knowledge is--with the help of certain theoretical devices--is the first step toward understanding the nature of language, and of the mind/brain.
One important point to note is that these intuitions we have, which are sometimes taken for granted, represent the creative aspect of our linguistic ability. For one thing, in spite of our vastly different experience, we seem to have no problem agreeing on our judgments about the examples presented above, some of which we clearly have not heard before. This means that what we know about our language must be rule-governed, and not the result of cumulative memorization. That is, as speakers of English, we must possess a set of rules that enable us--on the one hand--to produce and comprehend grammatical sentences and that prevent us--on the other--from producing or comprehending ungrammatical sentences. The rules make up the Grammar of English. Although we must have tacit knowledge of this grammar (or have "internalized" this grammar), we are not always able to explicitly state its content. As linguists, our task is to find out in explicit terms what those rules of English grammar are. We do this by the empirical scientific methods. We make hypotheses about the form and content of those rules on the basis of observations of speakers' intuitions, and further revise our hypotheses as we deal with larger and larger sets of data. Our hypotheses should reflect the generality of certain rules and their creativity, of course, and not a mere listing of facts. Thus the minimum requirement of a successful grammar is that it predicts the possible (actual and potential) occurrence of all and only the grammatical sentences of a given language.
While in the rest of this course we will be primarily dealing with English syntax, it is important to bear in mind that the study of English syntax is but a means to look into the general nature of language, and a facet of the inquiry into the nature of the human mind. Our study of syntax need not be based on English: study of any other language would contribute to the same goal. Following Chomsky, we assume that human beings are born with a language faculty in our brain (a geno-type structure called Universal Grammar, UG) which enables them, as they grow up, to acquire the grammar of one or more languages on the basis of their childhood experience. When one fully acquires the grammar of a particular language, we say that the geno-type structure has grown into a pheno-type, called English Grammar, or Japanese Grammar, etc. [This is the biological view of grammar. A grammar is a biological element much as any other human organ is a biological element.] Thus the ultimate task of the linguist, as a scientist, is to find out about the form and content of the UG. While an adequate grammar of a particular language must predict the possible occurrence of all and only the grammatical sentences of that language, the ultimate adequate UG must predict the possible forms of all and only the grammars of human languages (i.e. what constitutes a possible human grammar).
(1) Take the first stanza of Lewis Carroll's Jabberwocky, and
(a) identify the category that each word must or most probably belongs to,
(b) single out word(s) whose category seems difficult to identify,
(c) write a parody of this stanza with actual English words, using your own imagination.
Discuss and support your answers to (a) and (b) if you wish, for thoroughness.
(2) Based on your own intuition about English,
(a) create 5 new English words (new "words" that you are sure are not listed in any English dictionary, but are legitimate English coinages). Indicate why they are possible, though not occurring, English words, should the need arise to coin them. Exclude from your discussion any foreign words borrowed into English or foreign proper names.
(b) give five examples that cannot possibly be English words, and indicate why.
(3) Consider the sentences
(a) The quick fox jumps over the brown lazy dog.
(b) UCI students expressed dismay over the Regents' proposed fee hikes.
For each sentence,
(i) give an IC analysis of its structure,
(ii) assign a tree diagram representation to it, labeling each node with a category name.
(iii) translate the tree diagram into a format with labeled brackets.
(Do not worry if you are unsure whether some category labels are correct or not. This exercise is just for you to practice representing whatever intuition you already have.)