Part Six. Grammar versus usage.
Suppose you had created world-wide fame for yourself with proclamations about how you were going to do what no one had ever done before: explain how language works. Suppose you had dramatically swept away all the principles of the past and replaced them with new and revolutionary theories. Suppose you had dismissed the then-dominant theory of behaviorism and restored rationalism to its place as the driving force of this new science of linguistics. Suppose you had swept up along with you many people who were eager to be a part of the revolution, eager to take their places alongside you as the new mandarins.
And then suppose that you just couldn’t do it.
Suppose that nothing you tried actually worked out. Suppose that in no matter what direction you turned, your path was eventually blocked by inconsistencies and contradictions found in your theories.
What then? What do you, as the leader of an intellectual school of thought, do in order to keep your followers in thrall?
Simple enough: you do what anyone would do when faced with the unexplainable unknown: you turn to mysticism.
You say this: The brain contains many mysteries and therefore language works in ways that we simply don’t understand. You point to the fact that we intuitively know that some grammatical constructs are usable and some aren’t. You stop trying to explain rationally why this is so: that’s just the way the brain works.
From there you invent the original (and some would say astonishing) argument that communication never was the primary purpose of language in the first place! You say that the original purpose of language was to map and understand ideas inside the mind – in other words, a language of thought – and that communication came later – an afterthought in the sands of evolutionary time, you might say. You say that the language of communication is sometimes at odds with the language of thought, and therefore the former is what we call usage and the latter is the real grammar.
If you happen to be a nationally visible senior professor in the Chomskyan tradition, you publish a paper in a very prestigious journal of linguistics, in which you provide a couple of dozen examples which support this theory, evidence of the dichotomy between usage and grammar.
You hope that, by amassing a large number of examples, and claiming that none of them can be rationally explained, that even the less difficult examples, which seem questionable, will be accepted as supporting your theory.
You hope that no one reads your paper and tears your arguments to shreds. You hope that no one could possibly step forward and provide rational explanations for these mysteries of the brain, as to do so would show that all grammatical constructs are logical and that there are clear and understandable reasons why we say what we say. This, in turn, would destroy the dichotomy between grammar and usage and would be further proof of the decidedly anti-Chomskyan idea that language is created by humans using logical processes.
Unfortunately, your hopes do not materialize. I’ve read the paper. There are more holes in it than in a boxcar full of Swiss cheese. In fact, the paper may be unique in that every single supporting example given by the author – and I mean without exception – falls apart under close scrutiny! And this paper passed the peer review process of a top academic journal! Either that or it was given a free pass.
Let’s start with this example: the author of the paper, (I’ll call him Professor Fitz), asks this question: Isn’t it curious that English speakers don’t have a way to distinguish between second person singular and plural, even though such a form would be useful? Thus, something that would be good in terms of usage is absent in the grammar. Thus, the primary purpose of language cannot be communication.
Ahem. Is that really true, that which the professor claimed? Most Americans use the form you folks or you guys. Almost all Southerners use y’all and many Northerners use forms of youse. These look to me like ways of distinguishing singular from plural. When I made this point to an authority in academic linguistics, the response was “No, those don’t count because those are regionalisms.”
So let’s see if we’ve got this straight: the you/y’all distinction is used by eighty million inhabitants of the South – a region of the United States – but we must disregard that; on the other hand, the tu/vous distinction is used by a mere fifty million inhabitants of France – a region of Europe – but somehow that’s legitimate.
Clearly, English-speaking Americans see a need for a second person plural distinction and are creating various forms of it. The Chomskyans simply choose to ignore the facts which are right in front of their eyes, because the facts don’t fit with their theory.
(I should note that none of the forms noted above is a standard plural form, such as vous in French or Sie in German. Perhaps there is a simple reason that our use of second person plural is restricted to substandard forms. In languages such as German and French, the plural forms evolved to show not just plurality, but social distance or social stratification. Thus, in France one addresses individuals such as one’s boss as vous. In America, our democratic traditions influence our language, and so we do not use a standard plural form that can acquire secondary meanings of social distance or stratification. That’s why colloquial forms such as you guys and y’all and the like are used to indicate plurality but nothing else.) (Another example of our democratic traditions influencing language use is the current linguistico-diplomatic problem of his or her: Each student should put his or her pen on the desk.)
(Another question anyone should want to ask of Professor Fitz is this: Since so many languages such as Spanish, French, German and Russian do have standard second person plural forms, and English does not, and since Professor Fitz thinks that this lack in English is evidence of something universal at work inside the brain, does he mean to suggest that American brains work differently from the brains of Europeans?)
(And I can’t resist asking this question: Since Arabic has a special second person plural form available which is employed only when addressing females (though not much used anymore), would this be evidence that the brains of speakers of Arabic somehow work differently from the brains of both Americans and Europeans? Given the decidedly second-class status of women in many Arabic-speaking countries, would Professor Fitz simply dismiss out of hand the possibility that there may be a correlation between social structure (human interaction) and grammatical structure? Just asking.)
(One more aside and I’ll move on. According to the Linguistic Society of America’s Committee on Endangered Languages and Their Preservation, “Dozens of languages today have only one native speaker still living, and that person's death will mean the extinction of the language: It will no longer be spoken, or known, by anyone.”
Note that even though only one person speaks it (“To whom does he speak it?” being the unasked question), it’s still considered a language, at least by the LSA. However, if eighty million Southerners (a larger population than unified Germany) use y’all or practically the entire country uses either you guys or you folks – nope, sorry, that doesn’t meet the standard of bona fide language. That’s just a regionalism; that doesn’t count. Criminy!)
OK, let’s look at this next example from Professor Fitz: French is considered an SVO language, meaning that a subject noun acting on an object noun through a verb takes this standard form: Subject-Verb-Object: Marie voit la chaise. (Mary sees the chair.) However, it is common in French to use pronouns, and pronouns require this alternate form: Marie la voit. (Mary sees it.) Professor Fitz argues that, since a commonly used form (with the pronoun) has a different structure than the full form (with the object noun), then this shows that the commonly used form is distinct from the fuller, more grammatical form. Grammar versus usage, in other words.
Of course, this is an utterly ridiculous argument. Both forms are grammatical; neither one is more grammatical than the other. This so-called SVO designation is just a classification invented by humans (or, I should say, a human: the late Joseph Greenberg, to be precise.)
The fact that the pronoun structure doesn’t follow the same pattern as the noun object structure doesn’t prove anything. English is also an SVO language. We could say, Paul reads the book. Which is SVO. But we could also say: In the hammock lies Paul. Which is prepositional phrase-verb-noun, and is also perfectly grammatical.
Professor Fitz is not an obtuse fellow. I’ve read some of his works; he’s quite erudite. There is no way he could fail to see the fallacies in his arguments, as they are so obvious, at least in these cases. There are at least twenty more examples in his paper, most of which, however, are more complex than the above. Nevertheless, in each of his examples it is possible, if you search hard enough, to find a pattern or a logical explanation that shows why we use one form but do not use another. I know, because I’ve done it.
Let me show you one of the more difficult examples. I chose it for a specific reason, however; it has to do with a problem for which a satisfactory solution has never been published. The problem is called the Anti-Complement izer Effect (ACE). NYU’s Paul Postal, a very well known linguistics professor, first introduced me to the problem in a draft of his new book, Skeptical Linguistic Essays. The mystery around the ACE is illustrated this way:
Why is it that we think that Sentences (1) and (2) sound wrong but Sentence (3) seems fine?
(1) It is in these villages that we believe that can be found the best examples of this cuisine.
(2) It is in these villages that we believe that are contained the best examples of this cuisine.
(3) It is in these villages that we believe (that) the best examples of this cuisine can be found.
Many people have studied this problem and proposed solutions but no one has been acknowledged as the discoverer of the solution. I have proposed the following solution and can find no counterexamples:
(a) The word that (in bold) is called a complementizer because it introduces a complement;
(b) A complement is a complete thought and can be expressed as a sentence;
(c) As a complement is a sentence, it must take a form licensed by English syntax;
(d) If the complement is in a form licensed by English, then it can stand alone as a sentence;
(e) If it can’t stand alone as a sentence, then it can’t be preceded by a complementizer.
the best examples of this cuisine can be found.
from S3 can stand alone, but this from S1:
can be found the best examples of this cuisine.
and this from S2:
are contained the best examples of this cuisine.
cannot stand alone as English sentences, and thus cannot be preceded by complementizers.
And so if you take the complementizer that out of S2 and S3, you get these, which are fine:
(2) It is in these villages that we believe are contained the best examples of this cuisine.
(3) It is in these villages that we believe the best examples of this cuisine can be found.
(Note: in S3, the brackets around that indicate that the complementizer is optional and can be omitted. In S1 and S2, the complementizer is prohibited.)
To me, that’s a very neat and clear-cut explanation of the Anti-Complementizer Effect. Having resolved that, let’s now turn to the mystery which Professor Fitz discusses. Consider these sentences:
a. I wonder who you think likes George.
b. I wonder who you think George likes.
c. *I wonder who you think that likes George.
d. I wonder who you think that George likes.
(These particular sentences come to us from research done by Professor Wayne Cowart of the University of Southern Maine, as related in his fascinating book Experimental Syntax.)
Professor Fitz says, in effect, We know intuitively that Sentences a., b., and d. above are good, but we also somehow know that c. is bad. This must be the result of a formal grammar which is built into the brain, because there appears to be no logical explanation for these facts.
Ah, but there is a perfectly logical reason, Professor Fitz. The reason may be complicated, but that doesn’t keep it from being logical. It goes like this:
1. We can ask: Who likes George? and we can ask George likes who? Those are two separate questions, and it is the inversion of who and George which gives us the difference in meaning.
2. But look at that I wonder structure: the interrogative who must follow immediately after I wonder: I wonder who. Therefore we don’t want to move who to the end of the sentence: George likes who? We want to keep who toward the beginning.
3. There’s a way to do that: Who is it George likes?
4. Now let’s look at the you think part of the sentence. In Sentence d.
d. I wonder who you think that George likes.
the statement is really asking a question: You think that George likes who?
Now in questions which starts with you think, the word that is a complementizer, as that can be followed by this standalone sentence: George likes who?
Now let’s add I wonder to You think that George likes who?
We have to move who to the beginning and we get:
I wonder who you think that George likes.
which is Sentence d., and it’s fine. If we optionally leave the complementizer out, we get
Sentence b.: I wonder who you think George likes. which is also fine.
Now let’s look at Sentence a.: I wonder who you think likes George.
This is really asking: You think that who likes George?
Now let’s add I wonder to You think that who likes George?
And we get *I wonder who you think that likes George?
We have created Sentence c., which is bad. The reason is that moving who towards the beginning leaves likes George after that, and since likes George cannot stand alone as an English sentence, it cannot be preceded by a complementizer. In order to arrive at the good Sentence a., we must remove the complementizer:
a. I wonder who you think likes George.
Voila! A logical, albeit complicated means to debunk what the Chomskyans claim to be a mystery of the brain – a mystery which they believe supports the notion that true grammar follows its own rules, rules which are not influenced by anything so mundane as matters of usage or communication. Best of all, we use the solution to a heretofore unsolved problem (the Anti-Complementizer Effect) in order to do it.
Remember, you saw it here first.
Finally, let me note that the University of New Mexico linguistics professor Joan Bybee, recently elected president of the Linguistic Society of America, gave the keynote address at the society’s most recent national conference. In her address, she rejected the idea that grammar and usage were separate domains. Although she did not attempt a point-by-point refutation of Professor Fitz’s article, she wrote a paper which was full of common sense and which made a very good general case for her point of view. Her paper is called, not surprisingly, Grammar is usage and usage is grammar. It will be published in a forthcoming issue of Language.
Part Seven. Chom Ling and me: the saga continues.
Many readers of this article may recall my article “A Deaf Ear for Language,”in which I attacked the Chomskyans for trying to analyze this sentence:
Who will be easy for us to get his mother to talk to?
without their realizing that the grammatical form of the question should be:
Who will it be easy for us to get his mother to talk to?
My claim was immediately challenged by an anonymous and very capable Chomskyan linguist – I call him Chom Ling – and we got into a fast and furious debate using the comment pages of this website. In attempting to prove to the satisfaction of Chom Ling the ungrammaticality of the sentence in question, I did something rather ill-considered and amateurish: in order to prove that Sentence A was wrong, I tried to prove indirectly that a similar-but-not-quite-the-same structure, Sentence B, was wrong.
The editors of this website then published another article of mine, Chomsky’s Linguistics Refuted, in which I broadened the scope of my criticism. I attacked a major pillar of their theory – intuitivity – by which the Chomskyans say that whatever sounds correct to a speaker of a language is in fact grammatical. (See also Parts One and Two above.) I showed several examples of how intuitivity produces all kinds of problems for linguistic analysis, using example sentences such as Me and Larry went to the races., etc. I argued that one reason the Chomskyans have not been able to explain as much as they had expected was that they do not recognize a divide between grammatical and ungrammatical sentences. I also included the indirect proof described above.
Chom Ling contacted me via the website some weeks later, and pointed out some flaws in my indirect proof. I saw immediately that he was right, and went about doing what I should have attempted in the first place: a direct proof of the ungrammaticality of Sentence A. This I have done, and it is reproduced as a footnote at the end of this article. For those who do not wish to get into that much detail, I will give the highlights:
· There is a special structure in English which has been around for centuries, but which is used by only a tiny fraction of English speakers;
· The purpose of the structure is to allow one to be very grammatically precise in asking certain kinds of questions or in making certain types of statements;
· Although most people don’t use this structure and are in fact not aware of it, sentences constructed without it may sound OK, and yet may be technically ungrammatical;
· Chomskyans, as a matter of principle, concern themselves with what most people say or what sounds OK to most people (“We describe; we don’t prescribe.”) and thus, in the article in question, the author showed no evidence of his being aware of this special structure;
· Understanding why the special structure is useful, and knowing that it has been around for a long time, though little used, tells us something important about language, something that we can’t learn by studying what most people say, which is a Chomskyan precept.
I thank Chom Ling for pressing the point, as his persistence enabled me to go in and take a closer look at the problem in order to clarify my thinking. (I said “his persistence”, but in fact I have no evidence to suggest that the highly intelligent Chom Ling is a male. Unlike the president of a certain Ivy League university, I do not wish to suggest that scientific intelligence is more closely associated with one gender as opposed to another.)
Part Eight. Linguists gone wild.
If you have read this far, then you deserve a little diversionary treat. Take a look at this sentence:
Under the table was lying an elderly crocodile.*
What is the subject of this sentence? Is it:
A. An invisible subject?
B. The phrase an elderly crocodile? or,
C. The phrase under the table?
If you guessed B, you would be…wrong!!!
*I took this example from Paul Postal’s recent book, and I thank him for compiling the information.
You or I – as amateurs, I assume – might be inclined to say, Well, it’s the elderly crocodile that’s doing the lying, and was lying seems to be the verb, and under the table looks for all the world like a prepositional phrase. So what else could the right answer be but B?
Tut, tut. Child that you are, you will one day learn that it is simply not wise to make common sense assumptions when you’re dealing with modern theoretical linguistics.
Would you be surprised to know that a number of linguists believe that the correct answer is C? That’s right, the subject of the sentence is under the table. Seriously, you can stop laughing now.
You must be dying to know what the rationale is. Believe it or not, there is one. You’ll recall in Part Six above that I discussed the Anti-Complementizer Effect. Notice that the following sentence does not allow a complementizer (the bolded that):
(K) *It is these villages that we all believe that contain the best examples of this cuisine.
Although without that it is OK:
(K2) It is these villages that we all believe contain the best examples of this cuisine.
Nor can we allow a complementizer in (L), although (L2) is OK:
(L) *It is in these villages that we all believe that can be found the best examples of this cuisine.
(L2) It is in these villages that we all believe can be found the best examples of this cuisine.
Notice that Sentence K has a subject: these villages following It is…
Notice that Sentence L has a prepositional phrase in these villages following It is…
Thus, of the sentence containing the subject these villages and the sentence containing in these villages, both appear to be subject to the Anti-Complementizer Effect (ACE).
(Now you and I know that the determiner for the presence or absence of a complementizer has to do with whether the entity following the complementizer can stand alone as an English sentence or not.)
But that’s not the way others see it. Their reasoning goes like this:
· Since sentences containing subjects and prepositional phrase are subject to the ACE, that means that they have something in common;
· If they have something in common, then that means they must be fundamentally the same (!);
· Therefore, when prepositional phrases are in the subject position preceding a verb, the prepositional phrases are…subjects;
· Thus, under the table is a subject.
It’s completely cuckoo and you will be amazed to discover that PhDs at major universities are paid enormous salaries to drool on their keyboards and come up with stuff like this.
To continue reading, click here.