2.3 The computational system
2.3.8 Discovery of syntax: special features of the system Through study of syntax, linguists have now discovered specific prop-
18 c h i l d l a n g ua g e
If young children spoke only in static orders in repeated strings we would not say they had “acquired language.” In a sense, they must not only acquire the basic word order of the language, but acquire a “moving” system. The child in (2) above already knows this basic property of language (e.g., compare (2a) and (2b)).
2.3.8.2 Missing elements
Natural language productively hides its units. It involves numerous devices for reducing redundancy.11 Many sentence constituents are null (rep- resented as “Ø”).12 They are part of our knowledge, even though they are not realized physically in the sequence of sounds which we hear.
17. “We see a bee.
Now we see three Ø”
. . . .
“Ø Eat a snack” (Seuss, 1963)
In some languages, e.g., Chinese (18) or Spanish (19), null elements are more productive. (See Chomsky 1988a, 33; Huang 1984.)
18. Question: Zhangsan kanjian Lisi le ma?
(Did Zhangsan see Lisi?) Answer: Ø kanjian Ø le
Ø saw Ø (He saw him)
19. Ø Llega
Ø arrives (He/she/it arrives)
Children must acquire a system whose elements are in some sense free to
“disappear.”
2.3.8.3 Pronouns
Pronouns provide another way for natural language to reduce redun- dancy.
20. “Hewent into the tent”
“Theyyelp for help”
“That oneis my other brother” (Seuss, 1963) 21. “Mr Fox, sir,
I won’t doit.
I can’t sayit.
I won’t chewit” (Seuss, 1965)
11“Redundancy offers protection against damage, and might facilitate overcoming problems that are computational in nature” (Chomsky, 1991, 50); yet natural language productively reduces redundancy, e.g., through pronouns and various forms of ellipsis.
12Here and throughout, we will use the symbol “Ø” to represent the fact that an element exists in our representation of an expression, but is not spoken (phonetically realized). The term “null sites”
will be used to indicate their location.
20 c h i l d l a n g ua g e
Pronouns do not specify their reference and allow shifting reference. The duck and the mouse are confronted with this inAlice in Wonderland:
22. “said the mouse,” . . . “ . . . the patriotic archbishop of Canterbury, found it advisable . . .”
“Found what?” said the duck
“Found it,” the Mouse replied rather crossly: “of course you know what ‘it’
means”,
“I know what ‘it’ means well enough, when I find a thing,” said the Duck:
“it’s generally a frog, or a worm. The question is, what did the archbishop find?” (Carroll, 1998, 25)
The child in (2) has already acquired shifting reference with pronouns. (In fact (2)(a) through (2)(f) all involve pronouns.)
While “pronoun resolution” (determining the reference of a pronoun) remains one of the most challenging problems for formal machine-based computational approaches to “natural language processing” (NLP), it is one most naturally solved by anyone who knows a natural language. The language faculty of the human species appears to include particular facility for the special complex computation required by pronouns and null elements.
2.3.9 Knowing the impossible
The infinitely productive special combinatorial system which under- lies our language knowledge is infinitely constrained. Without ever having heard either the possible or the impossible sentences in (23)–(27), we know which are and which are not possible. So do children who acquire the English language.
Yet the number of “ungrammatical” constructions is infinite and so impossible to teach.
Constraints
Although infinite in capacity, combination and/or displacement in language is not always grammatical and thus not always possible. (23b) is not possible without changing the meaning of (23a) (in contrast to (15)). (23c)–(23d) in English are gibberish. Pronouns are not always possible with the same meanings, as in (24a) and (24b):
23. a. The poodle chased the beetle b. The beetle chased the poodle c. * Chased poodle the beetle the d. * Chased the poodle the beetle
24. a. The cat knew the boy likedhim(=the cat) b. The cat likedhim(not=the cat)
Constraints hold at every level (sounds, words, sentences), over all combinations of units. For example, we know which of the sound combinations in (25) are
possible English words or not.13We tacitly know these constraints, which explains why we do not speak gibberish.
25. ptak thole hlad plast sram mgla vlas flitch dnom rtut (Halle 1978, 294) We know that while the combinations of morphemes in (26) appear to create possible words in English, those in (27) do not.
26. a. overdose b. awesome c. downsize 27. a. *underdose
b. *bigsome c. *upsize
2.3.10 Finding the meaning
The formal computational system of language knowledge (syntax) must be integrated with other parts of human competence so that we, and children, cansay what we meanandmean what we say.
Lewis Carroll confronted the complexities of this mapping between the form of language and its meaning inAlice in Wonderland:
28. “Then you should say what you mean,” the March Hare went on.
“I do,” Alice hastily replied; “at least – at least I mean what I say – that’s the same thing, you know”.
“Not the same thing a bit!” said the Hatter. “Why, you might just as well say that ‘I see what I eat’ is the same thing as ‘I eat what I see’!” (1998, 64) Like us, Alice and the Mad Hatter must map the formal computational syntactic system of language to meaningand this meaning must be shared in order for communication to occur.
In part, the syntax of a sentence determines its meaning, as we can see simply by noting the difference between “I eat what I see” and “I see what I eat,” varying only word order. Acquisition of syntax is fundamental to children’s acquisition of meaning and avoidance of gibberish.
Not words alone. Not syntax alone.
As Alice discovered in her exchange with the Mad Hatter, words alone, even if we know their meaning and even if we organize them syntactically in sentences, are not enough.
29. “What a funny watch!” she remarked. “It tells the day of the month, and doesn’t tell what o’clock it is!”
“Why should it? “muttered the Hatter. “Does your watch tell you what year it is?”
13As Halle 1978 suggests, speakers of English generally recognize “thole”, “plast” and “flitch” as possible English words (294).
22 c h i l d l a n g ua g e
“Of course not,” Alice replied very readily: “but that’s because it stays the same year for such a long time together.”
“Which is just the case with mine,” said the Hatter.
Alice felt dreadfully puzzled. The Hatter’s remark seemed to her to have no sort of meaning in it, and yet it was certainly English. (65)
Alice most probably shares similar concepts related to the words (“watch”, “year”,
“time”) referred to, and she organizes her words in sentences, but she still does not fully share the meaning of her language with the Mad Hatter. We must also consult a theory of “pragmatics,” i.e., “use of language” to explain Alice’s challenge in Wonderland, and to fully comprehend what the child must acquire.