(I received many questions. They will be answered next week.)
Unisex English and generic they. My objections to the use of generic they aroused a good deal of interest, and after Andrew Sullivan reproduced part of my essay, it traveled widely through the media and blogosphere. Responses were also rather numerous. I’ll try to reduce my comments to a minimum. To begin with, it is advisable not to politicize the problem. My dislike of they in this role does not make me a reactionary; nor does my opponents’ defense of it turn them into freedom fighters. Also, the fact that I oppose generic they does not mean that I plead for the return of generic he. As for generic she, I find it to be unacceptable for two reasons. First, substituting the politically correct she for the condemned he is no remedy at all (if he is wrong, why is she right?). Second, she is very strongly “marked.” Centuries of use have made he at least partly neutral, but she inevitably makes us think of a woman. Saying something like when a criminal is sentenced, she should go to prison or when a professor walks into the auditorium, she always greets the students cannot be recommended.
Nowadays they is widely used in such contexts. Those who were born in the late sixties have been taught to say and write so and wonder what the fuss is about. The others have either conformed to this usage or avoid it. I was not out to fight generic they. To quote Fowler, who said so on another occasion, in the examples I used generic they is not wrong, it is merely repugnant. My goal was different. I took two sentences—“When a student comes to see me, I never make them wait” and “If a tenant is evicted, it does not mean they are a bad tenant”—and suggested that people had adopted this usage as a result of a deliberate attempt to cleanse English of sexism, but that in doing so, they had swallowed a medicine more dangerous than the disease. Consequently, the argument was not over the usage but over its origin. One authority after another declares that generic they has existed from time immemorial and that sentences like they were not a bad tenant or if your friend calls, I’ll tell them that you will call them back (you know their number, don’t you?) go back to early Modern or even Middle English and have been hallowed by the authority of “our best authors.” In my opinion, such sentences appeared only after some people, whose path to unisex English (a path paved, to be sure, by laudable intentions) enforced them on our contemporaries.
Specialists in the history of English are those who have read and analyzed hundreds (thousands) of pages written before roughly 1800. Those people never asserted that generic they had occurred in the old periods. The most passionate defenders of they did not study the history of English in any detail: at best, they read something about it and looked at a few samples. Their sources of inspiration are the OED and surveys derivative of it. The names of Shakespeare, Fielding, Thackeray, and a few others regularly circulate in the discussion. It is easy to see why. In the OED, the first famous author who is quoted as using they after an indefinite pronoun is Fielding. Let us now open Fowler’s Modern English Usage (the entry they, them, their, the first part). Naturally, he too speaks only about they referring to indefinite pronouns (someone, etc.) and words like individual, person, and each: “Undoubtedly, grammar rebels against their,” but “the OED quotes examples from Fielding, Goldsmith, Sydney Smith, Thackeray, and Bernard Shaw.” He reproduces the examples, which I have left out, and adds one of his own from Ruskin. It is this list, with occasional additions (Shakespeare, Scott, Shelley, Dickens), that is being constantly recycled. Every now and then, Samuel Johnson joins Fielding and others, but no confirmation of a wide use of generic they can be found in his dictionary. The research purporting to prove the antiquity of they were not a bad tenant resolves itself into copying a few lines from the OED and Fowler and finding other examples of the same type: they, them, their after someone, everyone, person, and so forth.
An irate correspondent deplores my meanness and lack of perspicuity and says that, according to the OED, nouns, not only pronouns, occur as the antecedents of they. It is a good idea to read the entire entry, rather than picking out one word from it. “Often used after a singular noun made universal by every, any, no, etc., or applicable to one of either sex” (= ‘he or she’).” The last part (“or applicable to one of either sex”) is unclear, but person, individual, and their ilk are meant, not tenant, student, or friend, because every animate noun, with the exception of boy, girl, man, woman, their close and distant synonyms, and words with a clear morphological structure (heroine, actress) are ambiguous as far as their referents are concerned: driver, artillerist, speculator—all of them, including boxer, wrestler, and soldier, can, at least nowadays, refer to both males and females.
The grandmother of another correspondent, a native of Texas, reportedly uses they in the constructions I mocked. The possibility that such usage might be widespread in dialects occurred to me, and that is why I consulted Joseph Wright’s The English Dialect Dictionary (the entry they: see the end of my original post) but found only examples like those given in the OED. It would be useful to know the old lady’s age, her parents’ dialect, and so on. Perhaps she simply repeats what she has been hearing for about 40 years. Anyway, no one believes that generic they came to Standard English from dialects. Still another correspondent remarks that even if generic they was introduced artificially, there may nothing wrong about it. I agree.
The situation is the same as with organic foods. Poisonous mushrooms are organic, but this fact does not increase their attractiveness, while homegrown champions may be tasty, even when not organic. Likewise, some natural changes in language are deplorable. A student who failed to submit her paper when it was due explained to me that its text was inaccessible to her because she had typed it up on her boyfriend’s computer, and “he, like, moved to Wisconsin, you know.” I know nothing about the young man and don’t like papers that disappear for such reasons, but this is not the point. No one taught that student to fill her speech with junk; she does it naturally. Conversely, in the 19th century, some Norwegians decided that their language was affected by Danish to such an extent that a new language going back to its rural roots is needed. They produced such a language. It was first called “country speech,” as opposed to “book speech,” and is now called New Norwegian. Today the Norwegians use their relatively limited resources translating books from one Norwegian to the other. This is a classic case of language planning. Some people bewail the results, others support them. Regardless of their attitudes, two Norwegian languages are here to stay. No one instructed the speakers of American English to change the rule of agreement and say the problem with your proposals are twofold and the information on whom will pay for this measure is absent. This is “natural” usage, and editors’ efforts to eradicate it will bring no fruit. I insist that generic they, unlike the problem… are and whom will pay (“wrong” but natural) was the product of language planning, while my opponents want to represent it as the outcome of language evolution. They want the construction I’ll speak to your friend when they call to have respectable ancestors. This privilege should be denied them.
So here is a challenge. Those who declare that generic they has evolved naturally are invited to send examples of the when a student comes to see me, I never make them wait and they are not a bad tenant type to this website. If they (real they) are specialists, they know the methods of linguistic research: find the earliest examples (Middle English, and so on), state in what area and social milieu they arose, explain how the phenomenon spread, and state when it was finally adopted by the speaking community. If pre-1970 examples (preferably, in respectable numbers) turn up, I will be the first to rejoice and concede defeat, but if the answer to my challenge happens to be silence, I will assume that such examples do not exist.
Gratuitous splitting. Comments on my blog devoted to the split infinitive, fortunately, lacked the militant spirit that informed the controversy over generic they. Several people defended splitting. Nor did I object to it, but I find thrusting to between not and the verb a bad idea (hence the title of my essay: “To Be or To Not Be”). I wondered where and why the splitting fad originated, but no one offered suggestions on this point. I keep thinking that infinitives should be split only when this operation is needed. In … the framework expected to easily pass in a referendum, the phrase to pass easily in a referendum would mean the same as to easily pass but “flow” better (or so it seems to me). The same is true of the youngest person to ever hold the U.S. Treasurer post (= the youngest person ever to hold), and especially of a decision… to not have political speakers of any persuasion (= not to have). The last two sentences occurred in the same article in a student newspaper. It matters little that, as a general rule, our young journalists are moderately literate and poorly educated, for theirs is the kingdom of the future. They were quick to jump on the splitting bandwagon. Soon they will grow up, and splitting will become the norm. It is up to our descendants to decide whether they will choose to have such a norm or to not have it, if I am allowed to put it this way.
English and the Indo-European family. Our regular correspondent and my frequent critic found fault with my statement (see the previous blog) that English has lost some of the most essential features of the Indo-European languages and is on the way of severing its ties with its family. He remarks that Indo-European is a genetic, rather than typological marker. To put it differently, once Indo-European, always Indo-European. By origin English is a Germanic and hence an Indo-European language. (How and when it became Indo-European is a special question.) But languages can also be classified from the point of view of their present day structure, and if we look at English, we will easily notice how far it has veered from its archetype. More than sixty years ago, N.S. Trubetzkoy, one of the greatest linguists of the 20th century, suggested that language families may emerge both because they happen to be offspring of protolanguages and as a result of long coexistence, as evidenced by a so-called language union in the Balkans (other language unions also exist). He applied his reasoning to Indo-European and arrived at nontrivial conclusions, but found few supporters. Yet his idea is fruitful and should not be dismissed out of hand. Other than that, I am not sure I understand how black children can be born to white Irish parents. As regards the ambiguity of phrases like small animal farm (a small farm or a farm for small animals?), compare free pregnancy test versus free trade test.
Anatoly Liberman is the author of Word Origins…And How We Know Them as well as An Analytic Dictionary of English Etymology: An Introduction. His column on word origins, The Oxford Etymologist, appears here, each Wednesday. Send your etymology question to firstname.lastname@example.org; he’ll do his best to avoid responding with “origin unknown.”