Oxford University Press's
Academic Insights for the Thinking World

Our shortest words continued: “of,” “both,” and (again) “if”

Last week, we looked at the history of the conjunction if, and it turned out that the Dutch for if is of. The fateful question asked “at dawn,” when “Scheherazade” had to stop her tale, was:  “Are English if and of related?”

Scheherazade at her best.

Unfortunately (or fortunately), the existence of family ties between if and of cannot be posited, because Engl. of is an impostor: it emerged as an unstressed variant of æf (the letter æ, called in Old English studies “ash,” had the value of a in Modern Engl. am). The historically true form can be seen only in the prefix æf. For instance, Old Engl. æfgrynde “abyss” is the exact analog of Modern German Abgrund (the same meaning), that is, “off (the) ground.” The German word shows that the form of our preposition had an a-like vowel, and indeed, æ is what we have found.

Engl. of, though a newcomer, felt quite comfortable in its new home and behaved like all social climbers: it ousted its “parent” æf and developed an emphatic form, namely off, which today can function even as a prefix (offhand, offshore). Although the tautological preposition off of surfaced in texts as early as the second half of the sixteenth century, it seems that its recent home has been the American continent. Jack London’s characters use off of all the time, while Dickens and Thackeray (both great lovers of slang and substandard English) hardly ever heard it. Of also sneaks in where no one needs it, as in outside of and alongside of. Conversely, it has been reduced to o’ in o’clock, tam o’ shanter, and the like. One cannot win all the time. Anyway, one tiny trace of af ~ æf may still be visible. The origin of Engl. after (with related forms elsewhere in Germanic) has not been settled; it may be æf– with a suffix of the comparative degree. Gothic also had aftuma “last,” and, unexpectedly, if-tuma “next, following,” with the variation a ~ i in the root (the product of the chaos whose politically correct name is ablaut).

The cognates of æf are not far to seek. Gothic, Old Icelandic, and Old Saxon had af; in Old High German aba, the last sound reminds us of ó in Greek apó– “away from.” English speakers know the Greek prefix from numerous borrowings, such as apostrophe, apotheosis, and so forth. In Germanic, stress was shifted, as always, to the first syllable, so that, not unexpectedly, the second vowel fell off, but in Slavic it remained where we find it in the protoform. Therefore, in that language group, the word’s first vowel was lost, which resulted in the appearance of the preposition and prefix po (for instance, po-russki “in Russian”); its Lithuanian congener is pa. Without some knowledge of historical linguistics it is hard to believe that German ab and Russian po are exact cognates.

I’ll leave it to serious students to explain why Greek has apó-, while German had ab (pay attention to the place of stress in Greek!), but those who know Swedish, Norwegian, or Danish will remember the preposition “on.” It goes back to what was upp á “up on; upon” in Old Norse. The vowel á, that is, ā (long a), became long o in the continental Scandinavian languages. In upp á, stress fell on the second syllable. Therefore, the first syllable was lost, and, as a result, we have the modern form . The process resembles the one described above for Slavic po. Woe to unstressed vowels!

Naturally, we still don’t know why the sound complex apó– or its ancestor meant “away from”; we hardly ever manage to reach such depths. (However, see the end of this post!). Since I have nothing else to say about the etymology of the preposition of, I will return to my previous post. In it, the Gothic word ibai “whether,” a cognate of Engl. if, was mentioned and presented as the compound i-bai. Bai, I stated, is either a particle or a word meaning “both.” Assuming that it means “both” (the more likely assumption), we obtain an elegant etymology of the conjunction if.

Engl. both in the form in which we today know it was borrowed from Scandinavian and consists of bo + th. Gothic had bai in the masculine and ba in the neuter. Old Icelandic báðir (read ð as th in Engl. the) consisted of – and a pronoun meaning “they” (thus, “both of them”). In Engl. both, –th is the stub of that old pronoun. If Gothic ibai really consisted of the pronominal root i-, glorified in the previous post, and a word for “both,” we begin to understand why it meant “whether”: the root pointed to a certain object, whereas “both” indicated the possibility of choosing between two options (both alternatives were possible!). Hence “whether,” “if,” doubt, hesitation, stipulation, and the rest mentioned a week ago.

The result is so inspiring that an additional comment on both may be justified. Gothic bai has a regular plural ending of an adjective, but, judging by its Sanskrit cognate u-bhāú, –ai was an innovation. The oldest Indo-European pronouns for the first and the second person had three numbers: singular, dual, and plural. The dual was used for “we two” and “you two.” Verbs, when they followed such pronouns, had corresponding endings. Almost nothing is left of the dual in Modern Germanic, but Engl. we is a relic of the plural, while Icelandic vit “we” continues the dual (originally we two). The ancient ending of the dual was –au. Of course, the word for “both” had to be dual! The choice suggested by it must have been between two objects, and whether referred to a similar binary choice, something like either—or.  This is the true beginning of if. To be sure, if is only what remains of the old form, but Gothic ibai shows how the history of that conjunction started. Incidentally, our alternating conjunctions were at one time and…and, or…or, not both…and and either…or.

Eight: The dual ideal

Since I found myself dealing with the dual and, considering the important fact that the posts for 12 July and 19 July 2017 were devoted to the numerals six and hundred respectively, I would like to dwell on the subject of the dual for some more time. The Gothic for “eight” was ahtau, an obvious cognate of Greek októ, with the ending preserved better in Gothic than in Classical Greek.  This ending suggests that eight was some kind of dual. But what was doubled? If it is true that at the beginning of civilization, fingers served as one of the tools for counting, we have in eight “twice four.” Two open palms, disregarding the thumbs? Perhaps so, but who can tell! It is no wonder that the etymology of eight remains a hotly debated question. Isn’t ten “twice five”? The Latin for “ten” was decem, and de– does make us think of “two” (compare Latin duo), but two of what? Two hands? Two units of some sort? In any case, ten is not a dual (though it could have been such!), and eight, most probably, is. Also both was once upon a time a dual, which leaves us, like the famed Buridan’s ass, facing a difficult choice: what if?

What if?

And now my last tale of this cycle is told, and the story teller is pardoned by the mollified sultan. The next post will broach an entirely different subject.

Image credits: (2) “Hands” by Erik F. Brandsborg, CC BY-SA 2.0 via Flickr (3) “Deliberations of Congress” Public Domain via Wikimedia Commons. Featured image credit and (1): “Sheherazade and Sultan Schariar” by Ferdinand Keller, Public Domain via Wikimedia Commons.

Recent Comments

There are currently no comments.

Leave a Comment

Your email address will not be published. Required fields are marked *