Oxford University Press's
Academic Insights for the Thinking World


It was not quite by chance that a week ago I embarked on a discussion of numerals (on June 19, the topic was nine). You are reading Post No. 700. Every hundred numbers, I congratulate myself on the progress of the blog. Oxford Etymologist came into existence in early March 2006, and, since a year consists of fifty-two weeks, it takes approximately two twelve months to produce a hundred illustrated essays. This past winter, there was a short break (for some time, the world languished without my contributions, though for no fault of mine); the schedule, it appears, may sometimes be disrupted. In any case, on July 12 and July 19, 2017, I wrote about the words six and hundred. Two summers from now, you may expect an essay on eight.

The number seven has not been neglected in people’s tales and scholarly thought. Don’t we have seven days in a week? The word sennight (that is, “seven nights”) used to mean “week.” In Middle English, the form ended in –nihte, and, according to the general rule, final -e was lost (apocopated). Sennight is hopelessly archaic, but fortnight “two weeks” is still a living word in British English. The Ancient Babylonians identified seven of their gods with the seven planets in the sky (apparently, the number included the sun and the moon, in addition to the five visible planets). All about such things can be easily found on the Internet, but few of our readers will have seen the periodical The Open Court (1887-1936). At one time, my team of volunteers screened the entire set for tidbits on etymology, and that is why I am aware of the article titled “Seven” by the editor Paul Carus (Vol. 15, 1901, 412-27); interesting reading. It too is now available on the Internet, and I will refrain from retelling it, but will add that in Pushkin’s (and Tchaikovsky’s) The Queen of Spades, the fateful cards are three, seven, and ace.

Seven was one of the cards that allowed the protagonist of “The Queen of Spades” to win. What about us, etymologists? Image credit: “‘The Queen of Spades,’ pg 102” by Paul Hardy, “The Strand Magazine.” Public Domain via Wikimedia Commons.

Etymologists are supposed to know why words mean what they do. Few words are more obscure than the names of our first ten numerals. In the post on six, I mentioned the well-known fact that in old days no one needed precise high numbers. Indeed, why should anyone want to mention 61 pigs or 93 horses? We have words like flock, herd, and a great number of others like them. Yet people have ten fingers, and, if we add ten toes to the count, words for the numerals from one to twenty will reflect tangible reality. There does not seem to be any disagreement that all primitive counting was and still is connected with the fingers.

Can you fine the 93rd horse in this herd? Image credit: “Horsemen leading a large number of horses in a ring for sale at a horse fair” by William Henry Simmons. after Rosa Bonheur. Public Domain via Wikimedia Commons.

I’ll borrow some facts from an article to which no one refers, but it is a curious piece and deserves a brief mention. The article bears the title “The Origin of the Names of Numerals.” It appeared in the renowned German philological journal Beiträge zur Kunde der indogermanischen Sprachen [“Contributions to the Study of the Indo-European Languages”], vol. 30, 1906, 223-65. Everything is unusual about it: the text is in English, it is very long, and the author is an American. The signature says: “Caroline T. Stewart. Columbia, Missouri.” From the ever-helpful Internet, I learned that she was a Michigan alumna and in 1905 had the rank of an assistant professor at Columbia, Missouri. A female professor of German in Missouri more than a hundred years ago was certainly not an ordinary phenomenon. She probably sent her entire MA thesis to Germany. The work is learned, but not original, though it contains a few suggestions that must have been her own. Adalbert Bezzenberger, the distinguished editor of the journal, hardly read the whole article. Yet he may have thought that his readers would be curious to know what philologists overseas wrote about Germanic and Indo-European linguistics, even though he could not be ignorant of the contributions by the great American sanskritologist W. Dwight Whitney (1827-1894). More probably, he appreciated Stewart’s footnotes about the systems of counting in the American aboriginal languages. I owe my information in the next paragraph to those footnotes.

William Dwight Whitney, a great American philologist. Image credit: “William Dwight Whitney” by unknown author. Public Domain via Wikimedia Commons.

In one of the indigenous dialects of West Washington, the root of the name for “finger” is also used for the word meaning “six.” Some people, it was observed, begin counting with the little finger of the left hand, proceed toward the thumb, which is 5, so that the next thumb will be 6, the first member of the new series—an important finger, perhaps even the finger. Elsewhere, for 5 the left wrist was touched, for 6 the left elbow, for 7 the left shoulder, for 8 the right shoulder, for 9 the left breast, and for 10 the sternum. Frequently, the same word is used for “five” and “hand” or for “five” and “fist.” If six may denote “the finger” and the beginning of the new series, it does not come as a surprise that the word for “seven” is sometimes related to the word for “two.”

We seem to be close to the solution of our own riddle, but, alas, we are not. Not a single English (or, for that matter, Indo-European) numeral resembles any word we need (finger, breast, shoulder, and so forth). The noun finger is also of uncertain origin, and I have doubts that it is akin to the German verb fangen “to catch”: no one “catches” anything with fingers. Fingers are nimble, so that the word finger may be cognate with many Germanic f-verbs meaning “to move briskly back and forth” (this etymology has been suggested, but has anyone noticed it?).

Seven pairs of all clean animals (and not only those) found refuge here. Image credit: “Noah’s Ark” by Edward Hicks, Philadelphia Museum of Art. Public Domain via Wikimedia Commons.

Another trouble with numerals is that they tend to influence one another. Four and five begin with the same consonant, and so do six and seven. The Gothic word for “eight” (ahtau) ends in au, which is a typical ending of the dual. (Indo-European had not only the singular and the plural but also the dual: “we two,” you two.” The verb had corresponding endings.) It follows that ahtau meant “twice four,” but this fact sheds no light on the origin of four. Finally, and this is the crux of the matter, in the past, numerals were occasionally borrowed from neighboring languages, because they played a prominent role in exchanging goods. There is no certainty that our words have ascertainable Indo-European roots. This disconcerting conclusion may be relevant to the origin of seven.

An often cited example of a borrowed numeral is Russian sorok “40.” Sorok (stress on the first syllable) does not sound like the transparent Slavic compounds for twenty, thirty, fifty, and the rest of them. The idea that sorok is a loanword makes sense. See also the post on eenie-meenie (“Catching someone by the toe”: January 10, 2018). In the Afro-Asiatic world, the word for “seven” resembles our seven, and the most recent theory has it that the Indo-European word is a borrowing. Where, in the Semitic group, the root has a transparent meaning, it denotes “middle finger; prominent finger on the hand.”

Tracing seven to a foreign source is acceptable, but there is a hitch. Though the scholarly literature on the Indo-European and Germanic seven is huge, most of it deals with the word’s phonetic form. The word’s irregularity and the absence of any plausible derivation make the suggestion of borrowing attractive. But Russian sorok is an exception in the series, while in Indo-European, the origin of all the first ten numerals is enigmatic. Should we then agree that all of them are borrowings? In the open court of etymology, the question remains unanswered.

Featured image: “Meeting the Seven Dwarves” by Loren Javier. CC BY-ND 2.0 via Flickr

Recent Comments

  1. Gavin Wraith

    You might be interested in the first chapter of The Book of Numbers by John H.Conway and Richard K.Guy (ISBN 0-387-97993-X). The late Professor Peter Hilton has also published remarks on the etymology of number words, but I cannot remember where.

  2. Stephen Goranson

    Peter Hilton wrote a Foreword, “Mathematics in Our Culture,” in the book, Mathematics: From the Birth of Numbers by Jan Gullberg (W.W. Norton, 1997).

  3. Yves Rehbein

    I find “sabath” and Proto-Semitic \*šabʿ pretty telling w.r.t. the seven-day-week and its weekend. Though I wouldn’t know if they had 7-days then, I don’t think so.

  4. Constantinos Ragazas

    Dear Yves,

    If there was no “seven day week” would there still be a “seven/Sabath” etymological connection?

    Did 7 exist before there was “Sabath”? And what people call 7 that now say “seven”?

    But here are some etymological connections you just can’t refuse:

    “one” from Greek “ena”

    “two” from Greek “dio”

    “three” from Greek “tria”

    “six” from Greek “exi”

    “eight” from Greek “octo”

    “nine” from Greek “enia”

    I’ll stop here! LOL!


  5. Yves Rehbein

    On the seventh “yom” god took a sabbathical to create a dam. I mean, come on!

    I found after a bit of theorizing that “evening” (e.g. Christmas Eve) is well comparable to “seven”. Which figures so sweetly with the time and again rejected comparison of “Acht” and “Nacht” (eight, night; octo, nox; But because of Sanskrit different medial velars are reconstructed).

    @Kostas related, of course, but the source of Hittite “2” in 1400 BC? Nevertheless, “ena”, “enia” seems to support the octal idea.

  6. Constantinos Ragazas


    As for the “…source of Hittite “2” in 1400 BC? “, one general comment.

    Whereas the I.E. root dā that appears in the Hittite ordinal dān has no meaning, the Greek “dio” has meaning in Greek! Meaning, “divided”. While Greek “ena” means “whole”.

    That said, quite likely the I.E. root “da” may in fact be the Greek “dio”. Bringing us back to the beginning of my original comment!


  7. Constantinos Ragazas

    Hello Yves,

    The supposed association of “seven” with “Sabath” gets even more hilarious! When you consider, in Greek, “Savato”/Sabath is Saturday. Which is a busy commercial working day. While Sunday (Kiriaki) is the day of rest and worship. And is the 1st day of the week.

    But, of course, Greek has a word for 7. And its not “seven” from “Sabath”!

    Greek “epta” (“seven”) through Latin “septem” may be etymologically closer to “seven” than “Sabath” is.


  8. Constantinos Ragazas

    Hello Yves,

    The supposed etymology of “seven” from “Sabath” gets even more hilarious! When you consider, in Greek, “Savato”/Sabath is Saturday. Which is a busy commercial working day. While Sunday (Kiriaki) is the day of rest and worship. And is the 1st day of the week.

    But, of course, Greek has a word for 7. And its not “seven” from “Sabath”!

    Greek “epta” (“seven”) through Latin “septem” may be etymologically closer to “seven” than “Sabath” is.


Comments are closed.