OUPblog > Language > Dictionaries & Lexicography > Back to the fishbowl (1): ‘Herring’

Back to the fishbowl (1): ‘Herring’

By Anatoly Liberman


The fish known as Clupea harengas has two main names. In the Scandinavian countries, it is called sild or something similar (this name made its way to Finland and Russia). In the lands where the West Germanic languages are spoken (English belongs to this group) the word is herring, also with several variants, for example, German Hering (the spelling Häring is quite obsolete), Dutch haring, and so forth. The rarely used English word sile “young herring” is a late adaptation of sild. The origin of both sild and herring is doubtful.

The salting of herring catch in Haringpakkerij, Amsterdam. Engraving. Early 17th century. Source: Wikimedia Commons.

As far as we can judge, people around the Baltic Sea consumed, cured, and traded in salted herring for millennia. In Old English, the word heringas (plural) appeared in the earliest glosses and rendered Latin sardinas. Another gloss gives hæringcas (æ has the value of Modern Engl. a in hat). The options facing an etymologist are familiar. Perhaps the word is Germanic and the fish was named for its appearance, habits, or habitat. Perhaps the people who settled on the shores of the Baltic Sea learned everything about this fish, including its name, from the extinct or assimilated natives. If such is the case, the sought-for etymology will never be discovered. But perhaps the word is not hidden in such depths and is a folk etymological alteration of an otherwise familiar name. Possibly, herring was in the past applied to a fish we only now call this; compare the gloss sardinas: heringas!

Herring don’t live in the Mediterranean Sea, so that its name could not have been learned from the Romans. However, as early as 1665, it was suggested that herring is an alteration of Latin halec. The reflexes (continuations) of halec occur widely in the living Romance languages (for example, Italian alece and Spanish haleche), and dialectal forms are numerous. The halec/herring etymology has been repeated in countless books and articles. However, the two forms are not so similar as to be easily confused. Besides, the earliest meaning of halec ~ hallec ~ halex was something like fish soup or fish sauce. Later it did refer to a fish species, but not necessarily the herring.

The English word is usually understood as stemming from hæring or hering with a long vowel in the root because in Middle English, spelling variants with ee occur. Besides, some Modern Frisian cognates indicate an ancient long vowel. We will assume that the traditional point of view is correct. At present, the vowel in herring is short due to later processes. But the oldest German forms also had short a. Therefore, a convincing etymology should not only explain “where the word came from”; it should also account for the variation long æ ~ short a. Such niceties didn’t interest pre-nineteenth-century scholars. Nor do they interest the modern public, but unless sound correspondences have been taken into account, all conclusions will have the solidity of castles in the air.

Vowels alternate in Germanic; compare write ~ wrote, bind ~ bound, drink ~ drank, steal ~ stole, give ~ gave, and many others. But they do so according to strict rules. Every series (like i ~ a ~ u in sing ~ sang ~ sung) resembles a railway track and the “trains” (vowels) are not allowed to change their routes. Every case of derailment shows that the pointsman/switchman (pointsperson?) — that is, the etymologist — made a mistake and paired incompatible forms. Despite such draconian rules, two views of derailment exist. Since accidents tend to recur (one can detect a certain pattern in them), some linguists recognize such deviations as legitimate. Others are stricter and believe that “wrong” alternations trace to an unrecorded, prehistoric language (“substrate”), in which different rules prevailed and from which Germanic speakers borrowed the words in question. The variation long æ ~ short a is not allowed in Germanic. Hence the predictable conclusion that herring is a “substrate word.” This, as noted above, may be true but since reference to a substrate is a dead end (the language is lost beyond recovery), relatively few scholars are in a hurry to move in that direction.

The main conjectures about the origin of herring are as follows. Herring swim in gigantic shoals and the fish name has been compared with the Germanic word for “army, troops,” such as German Heer, from heri (the word has not survived in English, but its root is extant in harbor, harangue, harry, and possibly harlot). In the past, it had a short root vowel, which makes it unfit for explaining the English word. By the way, some sources give Latin halec with long a. I am not sure why they do so; the best dictionaries cite halec with short a and long e. The etymology connecting herring with Heer (from heri) was proposed in the middle of the nineteenth century by the famous philologist Karl Müllenhoff and popularized by Friedrich Kluge, the author of the best German etymological dictionary. Apart from the difficulty with the vowel length, the problem with Müllenhoff’s guess is that heri never meant “shoal” (at least it has not turned up with this sense).

The other putative cognates of herring have long vowels. One of them is hair (Old Engl. hær ~ her, with long æ ~ e). Dutch haar also means “fishbone” and this is probably the reason the hair/herring etymology occurred to a Dutch scholar. However, the bones of the herring are neither too thin nor particularly fine, and persuasive parallels of fish being named for its “herring bone motif” have not been found. Yet this etymology keeps finding ever new supporters. One of them refers to folklore. In the tales of the world, the herring is often associated with the sun, which is usually depicted as a round face with the rays “radiating” (as they should) from the face; every ray is allegedly like a hair. Still another conjecture combines the elements of the previous two. Hair was believed to be a (or even the) source of strength, so it was surmised that hair in the name of the herring alludes to the great size of the shoal.

Who said that every herring drawn across our path is gray? Image by Lupo. Creative Commons License.

I will skip a fanciful etymology proposed by a distinguished scholar and only say that he attempted to find a Latin cognate of herring; this approach has little to recommend it. By contrast, the hypothesis by Otto Schrader, once a famous author, though now remembered only by specialists, deserves serious consideration. He compared herring and hoar (hoary). Engl. hoary, from har (with long a), means “gray-haired; grayish-white,” and so do its Scandinavian cognates. This link looks realistic; the herring is certainly “greyish-white.” If kipper, a word no less obscure than herring, is related to copper, with reference to the color of the male salmon, Schrader’s etymology finds additional confirmation, but no trust should be put in the support of one mystery by another. Fish often get their names with allusion to their color, spots, stripes, and so forth. For various reasons, similar things happen elsewhere in people’ interaction with the animal kingdom. From an etymological point of view, the bear is “a brown one” and the hare “a gray one” (like the herring?).

Unfortunately, now the forms with a short vowel remain in limbo. None of the proposed explanations shed light on this vexing problem. The influence of Latin halec has again been pressed into service, but it is better to forget that word. Then there is the substrate hypothesis, about which nothing can be said except for mentioning it. Fish names are often susceptible to taboo. The protoform of herring with a long vowel must have sounded approximately like hairingaz. Perhaps in some places it was changed deliberately to haringaz. Or perhaps in great trading centers, places in which pidgins are common, hairingaz was occasionally simplified to haringaz. More castles in the air.

Whatever the origin of herring, we should account not only for the root but also for the suffix. (It is a common fault of etymology that even seasoned scholars make do with a specious explanation of the stem and ignore the rest of the word.) The suffix -ing had a variety of meanings, but if one defines the most general one as “belonging to,” one won’t make a serious error. It is easy to understand why some linguists who thought of her- as referring to a shoal, glossed herring as “a fish in a shoal.” But even if we stay with “gray,” we will get an acceptable result, namely “a fish belonging to a gray multitude,” for in Old English -ing could be added to adjectives. The herring will emerge as a perfect partner of the whiting. Those are all shots in the dark. Yet I hope that, although we did not hit the target, once or twice we were close to it.

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 blog@oup.com; he’ll do his best to avoid responding with “origin unknown.”

Subscribe to Anatoly Liberman’s weekly etymology posts via email or RSS.
Subscribe to the OUPblog via email or RSS.
View more about this book on the

SHARE:

View more about this product on the

UK Website
USA Website
4 Responses to “Back to the fishbowl (1): ‘Herring’”
  1. [...] roots. It is rather the fanciful nature of the story that makes it almost impossible to believe. In the post on herring, I referred to the linguist who believes that herring meant the sun with rays all around its [...]

  2. [...] to resort to Etruscan. Unlike the pre-Germanic substrate about which something was said in the post on herring, the Etruscan language has not been completely lost. Several hundred Etruscan words, including a [...]

  3. Dear Mr. Liberman,

    the vowel difference -a- : -ê- can best be explained by the assumption of a collective Vrddhi-formation (herring as living in a swarm) – I would assume with the EWA a derivation of *(s)ker- ‘to jump, to dance’.

  4. [...] similar case is the fish name bass. (I am very happy to return to the fish bowl.) All its cognates have r in the middle: Dutch baars, German Barsch, and so forth. The word is [...]

Leave a Reply