By Anatoly Liberman
When we deal with the origin of ship and boat (the names of things pertaining to material culture), problems are almost predictable. Such words may have been borrowed from an unknown language (or from an attested language, but definitive proof of the connection is wanting) or coined in a way we are unable to reconstruct, but wife? Yet its etymology is no less obscure. My proposal will add to the existing stock of conjectures, and the future will show whether it has any chance of survival, let alone acceptance.
The few things that can be said about wife without hedging are as follows. In the past, it was pronounced wif, with the vowel as in Modern Engl. wee. It meant “woman,” not “female spouse,” as it still does in housewife, midwife, old wives’ tale, German Weib, and Dutch wijf. Very early, man(n) “person” was added to it, and by a series of phonetic changes wifman became woman. Old Engl. wif had cognates in German, Dutch, and Frisian. Old Icelandic wíf (í designates “long i,” the same vowel as in the Old English word) occurred in poetry, but whether it was native in Scandinavian or borrowed from English (a more probable option) is unclear. In any case, wif was not a common Germanic word, because it did not turn up in Gothic, a Germanic language, recorded in the fourth century CE. Nor is it a continuation of the main Indo-European word for “woman,” which we detect in gynecology and whose Germanic cognate is the now obsolete Engl. quean (quean is related to queen, but they are different words).
Wife, current in a large but limited area, seems to have been a term endowed with a specialized sense; otherwise, a cognate of Gothic qino “woman” (compare gyne-, above) would have satisfied the speakers. Equally problematic is the origin of bride, this time a common Germanic word. In dealing with woman, wife, and bride, we cannot remain in the sphere of “pure etymology,” for we have to investigate the family relations of past epochs and the exact meaning of kin terms. One word would designate a married woman, another a nubile woman, a third a bride, and so forth. In such cases manipulating roots and suffixes is insufficient, as the attempts to explain the derivation of wife show with depressing clarity. Hypotheses on the origin of wife are numerous, and the main obstacle confronting etymologists lies in the sphere of grammar, rather than semantics. Old Engl. wif (like Modern German Weib) is neuter. How could a noun meaning “woman” be neuter? No conjecture on the origin of wife is worth anything unless it can account for its grammatical gender.
But first here is a brief overview of the most popular theories. One book after another derives wife from the verb weave. Some people still support this derivation. However, it is indefensible from a phonetic point of view, and no one has been able to explain why the word for “weaver” should have been neuter. References to the supposed low status of old weavers are nonsense. Then there is Gothic (bi)waibjan “surround, encompass; clothe, wrap” (bi- is a prefix). Its cognates usually mean “swing, sway, vacillate,” as seen in Engl. waver. Biwaibjan and its congeners gave rise to another well-known etymology of wife. Allegedly, the sought-for link between wife and clothe was the veil. “Wife,” according to this reconstruction, meant “a veiled bride,” because the veiling of the bride was customary among all Western Indo-Europeans. However, apart from numerous semantic complications, which I will skip, it remains a puzzle how the line between “bride” and “woman” was crossed (no society, and hence no language, confuses these concepts) and why a female person about to marry, even if veiled, acquired the neuter gender in Germanic. Other suggestions along the same lines were no more persuasive. Instead of “veil,” various pieces of a woman’s apparel were named, but the basic idea remained: “from clothes to person”, as in he chases every skirt. However, there is hardly a single solid example of a word like skirt, apron, or bonnet turning by metonymy into an everyday name for “girl” or “woman.”
As could be expected, some people hoped to find the etymon of wife in a word for the woman’s genitals. A neuter noun holding out some promise turned up only in Tocharian, which is not good for this etymology, because wife, as noted, had limited currency even in Germanic. In Tocharian B it sounded kwipe and meant “shame place,” with reference to “penis,” whereas Tocharian A kip meant “mother’s shame body,” so either “vulva” or “womb.” Secure Tocharian cognates of even Common Germanic words are not too many, and, if the Tocharian noun were related to the protoform of wife, it is almost unimaginable that this word would not have turned up somewhere between Asia Minor and Medieval Germania. Also, as can be seen, neither kwipe nor kip meant directly “woman’s genitals.” However, this etymology, like all the previous ones, found a few distinguished supporters.
I will pass by other, even less convincing, conjectures and come to my own proposal. Discrepancies between the grammatical gender of the word and the sex of the person it designates are not uncommon, and several other examples of neuter nouns for “woman” exist. Each of them needs a detailed explanation. Here only one fact should be mentioned. In all the old Indo-European languages the form of the feminine singular coincided with that of the neuter plural. This circumstance poses interesting and complicated questions about the origin of the grammatical gender and relations between a group (for which the neuter plural is natural) and an individual woman. In any case, the path for a collective plural to a singular, either feminine or masculine, has been attested more than once. An anthologized example is god. Old Germanic had only the neuter plural (gods). The masculine noun appeared after Germanic-speakers were converted to Christianity.
Among the Old Scandinavian goddesses, we find Sif. Her name, derived from Indo-European si-bh, is related to Engl. sib and Latin su-us “one’s own.” Sif must have been the patroness of family ties. The only recorded myth in which she plays a visible role, points to fertility, rather than affinity by marriage, but the concepts of family and fertility are close. I compared Sib and the personal pronoun we. The protoform of we was wis (with “long i, that is, wees, if spelled in today’s English); -s was an ending. I think that Old Germanic wibh, the protoform of wife, was wi-bh a formation parallel to sibh. If I am right, sibh meant “all the people related by marriage,” while wibh referred to a group tracing its origin to the same woman. It was a word like y’all. Wibh, as I see it, had to be neuter, because it was the name of a community whose members descended or believed that they had descended from the same woman. It included both males and females, and in Germanic, when a pronoun like they covered “mixed company,” the form was always neuter (John and Jack needed the masculine they, Betty and Mary would be covered by the feminine they, whereas Jack and Jill required the neuter they). As time went on, the word meaning “we, descendants of one woman” came to mean “woman.” Wife emerged as a term of social relations, but the old grammatical gender remained. The old Indo-European word for “woman” (preserved by Engl. quean) also survived, but it narrowed its sphere of application and came to denote “woman in her biological (child bearing) function.”
Those who have trouble believing that the same word can refer to a group and to an individual should recall Engl. youth “young people” and youth “a young man” or people and a people, let alone one sheep ~ many sheep, whatever the causes of this usage may be. Apparently, the distinction between a woman’s role as the founder of a clan and her role of a potential mother was not universal, because even in Germanic it was limited to a certain area. Nor have we retained it: woman serves both purposes equally well, and quean, to the extent that it is still used outside some rural dialects, means “slut.” Somebody may say that Germanic wib has not been attested in its collective meaning. Quite so. If this sense had continued into the literary epoch or lingered in some archaic dialect, my etymology would have been offered in the eighteenth century at the latest.
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 him care of firstname.lastname@example.org; he’ll do his best to avoid responding with “origin unknown.”