Oxford University Press's
Academic Insights for the Thinking World



Soon after the previous gleanings (February 26, 2020) were posted, a correspondent asked me to clarify the situation with the “prefix” br– in breath and bring (see the post on breath for January 22, 2020). I mentioned this mysterious prefix in connection with Henry Cecil Wyld, who accepted its existence in bring but doubted its validity in breath. From a historical point of view, we have two different components, even if both go back to Indo-European bhrē-. James A. H. Murray thought that br- in breath is a remnant of the root meaning “burn,” as in breed ~ brood, while br– in bring traces allegedly to the zero grade of the verb bear (zero grade is a term of ablaut; in this case, no vowel stands between b and r in br-; hence, “zero”); so Wyld, though, as we will see, the idea was not his. By contrast, in the full grade, as in bear, from Old Engl. beran, the syllable is supported by a vowel. Thus, bhr-1 and bhr-2, if those entities are more that figments of etymologists’ imagination (as they may well be), have different histories. But one feature unites them: if br-eath and br-ing consist of two parts each, both are blends, like Lewis Carrol’s galumph (gallop + triumphant) or Oxbridge (Oxford + Cambridge).

In the recorded texts of the oldest Indo-European languages, blends turn up most rarely, but in 1937 the German scholar Rudolf Blümel cited a few good examples in Classical Greek, and in 2019 Ryan Seaberg (the University of Minnesota) defended a dissertation on blends in Greek and Latin. Similar “portmanteau words,” typical of colloquial speech, must also have existed in the remotest past, for words have always tended to merge, and there is no great difference between blending and compounding.

Karl Brugmann, 1849-1919. From Wikimedia Commons.

It may be instructive to throw a quick look at the etymology of some verbs having nearly the same meaning as bring. Bring can be synonymous with take, as when we TAKE an object away from our location but BRING it to the place where we are. Engl. take is a borrowing from Scandinavian (German nehmen is the verb, whose cognate English has lost; today, only its semi-obliterated traces can be detected in nimble and numb). Scandinavian taka meant “seize, receive,” etc. Then there is carry, another borrowing, this time from Old French; its main sense must have been “to move.” Fetch is English; it is a cognate of German fassen “to seize, grasp.” Conduct, convey, and transport will add nothing new to what we have seen above. The origin of German tragen “to carry” (cognate with Engl. drag ~ draw) is obscure. Those examples will suffice to show that, though the basic meaning of bring is rather vague, it usually refers to getting hold of and moving an object. Verb-adverb collocations like bring up and bring out show how pliable the verb bring is.

Misery and luxury: two images of overexposure. First image is Defoe in the Pillory. CC BY 4.0 from Wellcome Collection. Second image, dandy, public domain from Wikimedia Commons. Image has been cropped to fit.

The origin of bring remains a riddle. The earliest etymologists of English, German, and Dutch had nothing to say about it. Later dictionaries cited the unquestionable congeners (cognates) and stopped there. Bring has related forms in all the Germanic languages (including Gothic, a dead language, known from a fourth-century translation of the New Testament), except, for some reason, Old Norse. A breakthrough happened in 1901, when Karl Brugmann, a famous German scholar, offered an imaginative etymology of this verb. But before discussing his hypothesis, I would like to mention the musings of Ludwig Laistner, another German philologist. He is remembered for his works on myths, but almost (or even entirely) forgotten as an etymologist. Although he hardly discovered the origin of the verb bring, his suggestions going back to 1888 are worth recalling.

The Gothic verb meant not only to “to bring” but also “to get something done,” approximately like Modern German vollbringen (voll “full”). Elsewhere in Germanic, bringan, the oldest recorded form of bring, must have given speakers grief. There are pairs like Engl. sit and set. Set is a so-called causative verb; it means “to make someone or something ‘sit’.” Bringan also had a causative twin, namely brengan (apparently, from brangjan), which, surprisingly, meant the same as bringan! As a result, the Modern Dutch for bring is brengen. Laistner emphasized the reference of the Germanic verb to accomplishment, regardless of whether the result was good or bad. He cited Gothic –praggan (gg = ng) “to oppress” and German Pranger “pillory,” as opposed to Prunk “splendor” (!). All of them seem to go back to the idea of bringing things to the surface. In those words, initial p alternated with b. Thus, Middle High German brunken meant “to show, expose,” and ge-brunkel can be glossed as “sheen of armor.” In the history of German, b and p constantly played leapfrog, but a Gothic non-borrowed word beginning with p is a great rarity, even an anomaly. As far as I know, since Laistner’s days, no one has thought of bring in the context of –praggan, Prunk, brunken, gebrunkel, and the rest, and yet his observations deserve attention. It seems that the ancient meaning of bring was indeed “to expose,” rather than or at least in addition to, “move an object from place to place.”

Several things are “wrong” with bring(an). First, the alternation b ~ p in the words that may be related to it. Second, its local spread. Perhaps some Celtic forms are akin to bring, but their affinity is open to doubt. Bring has no unquestionable Indo-European relatives and did not make it to Old Scandinavian. Third, the presence of a semantically redundant causative verb. And finally, its conjugation. All old verbs that rhyme with bring are strong, that is, their forms are governed by ablaut: spring—sprang—sprung, sing—sang—sung, etc., while bring is weak (bring—brought—brought, like seek—sought—sought). We seem to be dealing with an anomaly and a linguistic misfit.

Even if brunken is discounted as a false cognate, brengen, from brangjan, shows that bring took part in the ablaut game. The past form brang did occur, but it is usually explained away as an analogical formation. Was it really? In any case, weak verbs were not supposed to have vowel alternations by ablaut. The old verb meaning “to carry (from place to place)” is bear—bore—borne. Who needed bringan? As usual in such cases, some etymologists suggested borrowing from a substrate language. Yet it is better to stay away from the mysterious substrate, even though bring(an) does look like an intruder, for the reason bring made its way into Germanic remains unsolved. Are we dealing with Old Germanic slang? Perhaps, though reference to slang is not an etymology.

I am now returning to Brugmann. He suggested than bringan was a blend of two roots: bher-, as in bear “to carry” and enké– or (e)nek– “to reach” (the second root has been preserved in Greek and can be detected in Engl. e-nough, from ge-nōg ~ ge-nōh). If such was the formation of bringan, this blend was, most probably, slang. Brugmann’s reconstruction does not account for the meaning “to expose,” but at least it makes clear why the verb has no indubitable cognates (very Old Germanic-Celtic slang is quite probable, but, as noted, the Celtic forms are not watertight) and why this verb is anomalous in so many respects. Another etymology exists. Bringan might be beran “to bear” with n inserted into the root: compare Engl. sting and Latin (in)stīgare “to instigate.” However, this scenario is less likely: too many features in the history of bring require an explanation, so that the idea of a slangy blend looks more promising.

Such is the shortest history of br- of bring and breath: not a solution, but at least a reasonable hypothesis. There is some irony in the fact that English has the phrase to bring to bear. Brugmann did not speak English, but he could read it, and, if he had known the idiom, he would have appreciated it as an unexpected tribute to his sagacity.

A weight brought to bear. Public domain from Pixabay.

Feature credit image: Small mouth black bass. Public domain from Flickr.

Recent Comments

  1. Stan Smith

    Bring – brung

    I heard ‘brung’ frequently in my South Lancashire childhood, as both past participle and verb. (N W Midlands dialect). I seem to remember it in medieval poetry – Gawain and the Green Knight? – but this may be a false memory.

  2. Jean

    Can you comment on any other Italian words or place names that relied on the accented syllables surviving, as is shown in “Milan” derived from “Mediolanum”?

  3. Jevgēnijs Kaktiņš

    It seems Latvian braukt – to drive by transport (nowadays) participated in the game (its meaning switched some day) as Lithuanian word has completely different meaning I can’t even translate. wiktionary attributes some other ancient Baltic words in the same company (as branktas for whippletree)

  4. Constantinos Ragazas

    Anatoly you write,

    “It seems that the ancient meaning of bring was indeed “to expose,” …”

    If “to expose” means “to find”, there is an interesting association here of “bring” with the Greek “βρηκα” (find). Which has the same ancient Greek root as with “εφευρεση” (“discovery”).

    All this perfectly in agreement with what you are arguing here. Since to “find” means to “bring” to light which means to “expose” which means to “discover”.

    It all ties nicely together, if we go back to Greek.


    “The origin of German tragen “to carry” (cognate with Engl. drag ~ draw) is obscure.”

    Not too obscure! If we consider the Greek “τραβώ” (“draw, drag, pull”). As clear an origin for English “draw” and German “tragen” as you can find!


Comments are closed.