23 September 2015

Nucg Nucg, Winc Winc: The Anglo-Saxon Dairy Business

Those of my visitors who know something about Old English poetry may have realised that the link between the F-word and churning butter (see the previous post) is not just etymological – it’s a literary allusion.  Among the famous Anglo-Saxon riddles preserved in the Exeter Book we find the following one (Riddle 54):

Hyse cwom gangan,    þær he hie wisse
stondan in wincsele,    stop feorran to,
hror hægstealdmon,    hof his agen
hrægl hondum up,    <hrand> under gyrdels
hyre stondendre    stiþes nathwæt,
worhte his willan;    wagedan buta.
Þegn onnette,    wæs þragum nyt
tillic esne,    teorode hwæþre
æt stunda gehwam    strong ær þon <hio>,
werig þæs weorces.    Hyre weaxan ongon
under gyrdelse    þæt oft gode men
ferðþum freogað   ond mid feo bicgað.

An Anglo-Saxon churn lid, with the Freudian hole
The Exeter Book (written more than one thousand years ago) is the largest extant anthology of Old English poetry. It contains diverse stuff, from solemn religious and allegorical poems, saints’ lives, elegies and fragments of heroic legends to comic, somewhat naughty, light compositions, such as Riddle 54. There are as many as 96 Old English riddles in the manusctipt (the genre is hardly documented in any other source). Many of them have very serious religious solutions, but certainly not this one. Good translations of the riddles are hard to get by. Much is lost in translation, and humour is usually the first victim. A specialist can always enjoy the original, but for the sake of those whose Old English is not very fluent I’m going to offer my own translation, for what it’s worth. At least it isn’t a horrible mistranslation (some others are) and it tries to capture the spirit of the original. I also hope it isn’t too stilted (for a piece of Old English verse).

Some things are practically untranslatable. For example, Old English had grammatical gender, and the use of feminine personal pronouns (corresponding to Modern English she and her) doesn’t mean that the pronoun indicates a female human being. It can refer to any object whose Old English name is a feminine noun (e.g. tunge ‘tongue’, bōc ‘book’,  duru ‘door’, etc.). It may suggest a woman, but since the alternative possibility is also probable, the suggestion is much weaker than in Modern English. This subtle ambiguity would be lost completely if she were replaced by it, so I let it stay. Just remember that in Modern English not only ships but also some tools and utensils can be conventionally personified by their users and referred to as “she”. It isn’t quite the same thing as Old English grammatical gender, but must suffice to justify my artistic licence.

Another problem is that Old English is a dead language and its written record if far from perfect. The words in angle brackets represent editorial emendations in places where the text seems to be corrupt. The first of the restored forms, <hrand> actually reads rand in the manuscript, but this can’t be the word intended by the poet. The rules of Old English poetic alliteration demand something beginning with h in the first stressed position of the second half of the line. The most likely emendation is hrand. Unfortunately, such a word-form does not occur anywhere else in the entire Old English text corpus. The context requires a verb in the past tense here. A past tense like hrand presupposes the infinitive *hrindan, past tense plural *hrundon, past participle *hrunden, etc. But what might they mean? Not only is the verb otherwise unknown from Old English; it has left no Middle of Modern English descendants either. To use a technical Greek term, it’s a hapax legomenon, a word appearing only once.

There’s nothing wrong with being a hapax. It’s the inevitable consequence of the fact that words have wildly different frequencies of use (a common motif in my blog posts). In fact, in any large corpus of texts at least about 40% of the words (types, not tokens) occur only once. The same is true of Old English: more than half of the entries in any more-or-less complete Old English dictionary occur only once or twice in the surviving texts. So hrand is not anything unusual, just a little enigmatic.

What about possible cognates in other Germanic languages? We have Old Icelandic hrinda (past tense hratt < *hrant < *hrand) whose precise meaning is known: ‘push, hurl down’ and, figuratively, ‘launch’ or ‘expel, get rid of’ (the verb has survived in Modern Icelandic and Faroese). The literal meaning roughly fits the context of Riddle 54. Most Modern English translations use thrust; I prefer shove because of its greater semantic overlap with Scandinavian hrinda, and also for the sake of alliteration. Last but not least, shove is less dignified than push or thrust, and has the kind of colloquial vigour they lack, which is an advantage in this case. All right, I’ve never tried it before, so here goes!*)

A lad came walking    to where, as he knew,
she stood in a corner;    stepped in from afar,
a brisk bachelor,    tucked up his own
shirt with his hands,    shoved under the girdle
of the one standing    a stout thingumajig
and worked his will;    both rocked back and forth.
The servant quickened up:    at times he was of use,
a handy workman,    he grew weaker though
with every stroke,    strenghtless too soon,
weary from work.    There began to form
under her girdle    that which good men often
dearly desire    and procure with money.

And the solution is yes, yes, you’ve guessed correctly!  a butter churn, that is OE ċyrn. By the way, this word occurs three times in Old English texts: once as cyrin (sg.), once as cyrne (pl.), and once as cirm (misspelt by the scribe). As you can see even the citation forms that we use for convenience represent “Standard Old English” imposed by modern dictionary editors rather than the actual language of the manuscripts.

An early 20th-century postcard
Needless to point out, ĊYRN [wink wink, nudge nudge, say no more, say no more] is the “formal” solution of the riddle. The informal one is as obvious to us as it was to any Anglo-Saxon audience in the tenth century. Other ambiguous riddles in the Exeter Book exploit the same risqué ambiguity: the alternative interpretation is invariably bawdy. Their innuendo-laden humour may be crude, but it still appeals to the modern reader. For the survival of the whole collection we are indebted to Leofric, Bishop of Exeter, a well-educated bibliophile, who died in 1072, bequeathing his impressive manuscript collection to Exeter Cathedral. He apparently did not regard the riddles as subversive enough to be denied the shelter of the cathedral library. Riddle 54 helps us to understand why, back in 1290, a chap from Ipswich, presumably a local dairyman, was called Simon Fukkebotere. It offers us a glimpse into the secret world of naughty associations that existed in the minds of Anglo-Saxon scribes and their audience (and still exist in ours), so we are not making things up when we hypothesise that the original meaning of fuck was ‘strike repeatedly’. Who knows, perhaps the speakers of Old English could use the same word for churning and, with less innocent intent, for [know what I mean? nudge nudge] the other thing.

*) ... said the actress to the bishop.


  1. This fits the claim that Muckefuck is so called because it's apparently stirred a lot, rather than (as Wikipedia says) because it might as well be made from rotting wood. Too bad I can't remember my source.

    and once as cirm (misspelt by the scribe)

    Y confused with i in OE already? Where is that manuscript from?

  2. I and y are normal Late Old English developments of the rather short-lived diphthong (Early West Saxon ie, Anglian io) which resulted (in this word) from the breaking of Pre-English *i before preconsonantal *r. Early WS *ċiern is the expected reflex of Proto-West Germanic *kirnjō. The *i in the first syllable is the Common Germanic outcome of *e affected by the assimilatory influence of *j in the next syllable ("early umlaut'). There was also a competing West Germanic variant -- a simple feminine stem, *kernō. In Scandinavian, the 'churn' word was a weak (n-stem) feminine -- either *kernōn- or *kirnjōn- (the latter in Old Icelandic kirna).

    In cirm, the m is a scribal error for in.

    1. P.S. A small correction: the normal LWS development of EWS ie was y. But in EWS the spelling ie alternated orthographically with i, which probably doesn't indicate a full merger (since it would have had to be "unmerged" in LWS). Other dialects (Kentish, Anglian) retained "broken" io, which by the end of the Old English period fell together with eo.

  3. Just as a matter of curiosity, the Basque verb jo shows a similar semantical relationship: it means both 'fuck' and 'strike' since earky attestations in the 16th c.

    1. Thanks. It is also worth remarking that in Slavic, which has the aspectual contrast between perfective (punctual) and imperfective (durative, iterative) verbs, imperfective verbs meaning 'fuck' can be transformed into secondary perfectives meaning 'strike'. For example, Polish jebać 'fuck' (a verb as old as the hills, cf. Vedic yábhati) forms the perfective jebnąć 'hit hard'. It inherits the vulgar character of its morphological base, but the meaning is not sexual. Also the more recent (and slightly less obscene) synonym of jebać, pierdolić (which can also mean 'talk bullshit'), yields the perfective pierdolnąć with the same "asexually vulgar" meaning as that of jebnąć.

      I love the explanation of futuo in A Latin Dictionary (Lewis & Short): to have connection with a female (rare).

    2. As we know, jo's primary meaning is "hit," "strike" with a secondary sense of "to fuck" (mostly from the male point of copulation, compare jokhatu, which Michelena noted in older times was strictly used for birds). I am guessing sexual sense derives from the older phrase larrua jo "to fuck" (lit. "beat the hide," "strike the skin/pelt")

    3. I'm sorry, Patrick, but I don't agree with you. If we look at the old attestations (and regardless what modern dictionaries say), from the very first attestations of jo it had a wide range of meanings such as 'to wound', 'to hit', 'to collide', 'to head/go/reach somewhere'... but also (and, which is important, in all the dialects) 'to fuck'. On the contrary, larrua jo is a dialectally restricted and diachronically later attested collocation.
      Moreover, most of the first attestation of jo are, indeed, with the sexual meaning (in different dialects, both in the literary language and in the natural speech attested in court records).
      Regarding jok(h)a (-tu is the participial suffix), which is not general, its sexual connotations are (even more) dialectally restricted, and clearly secondary (and later attested, as opposed to the underived jo). On the other hand, its composition resembles strikingly the Slavic aspectual contrast mentioned by Piotr: jo (the verb stem) + ka (adverbial suffix of repetition).

  4. https://38.media.tumblr.com/cd82f8068a5f02c06b2f9222b741dbc0/tumblr_moo2zeIOlw1qlzduwo1_500.gif

  5. Here's yet another article on the similarity of "mom" and "dad" words across the world. Would you (Piotr) agree with what the author wrote about the IE word for "short" developing into "upper arm" and the English word "merry"?'


    In Proto-Indo-European, the word mregh meant “short.” The Greeks’ version of that word came to refer to the upper arm, which is short, while in Latin it referred to a pastry that looked like crossed arms; the term then passed into French referring not to arms but shoulder straps. All of those words seeped into English later, such that what started as a word meaning “short” became “brachial” (from Greek), “pretzel” (the crossed arms, from Latin), and “bra” (“shoulder strap” became brassiere). The most direct descendant of mregh in English is “merry,” of all things. That which is short is often sweet, such that the word came to mean “short and sweet” and, eventually, just sweet—merry, that is.


    1. It's all correct, more or less. Personally, I prefer to be more precise and not to talk about "words" when I really mean "roots". The comparatively reconstructible adjective is *mr̥ǵʰ-ú- 'short'. The comparative form was *mréǵʰ-j(o)s- 'shorter'. In Greek we have a regular reflex of the basic adjective, brakʰús 'short'; same in Avestan, mərəzu, and even Germanic: OHG murg 'short', Goth. ga-maurgjan 'shorten'. In Latin, the full grade of the comparative was generalised, and the adjective was moved to the i-declension (together with other u-adjectives): *mréǵʰ-w-i- > Lat. brevis. Note that we still have no semantic change. All these words still mean 'short' or 'brief'. Of course in all these languages the adjective could form derivatives, and the very purpose of using the derivational system is to get new words with a modified meaning.

      Hence developments like Gk. brakʰī́ōn 'the shorter (part)' --> 'the upper arm' (but later also 'arm' and 'forearm', thus e.g. in Aristotle), borrowed into Latin as bracchium 'forearm' (but also 'arm', from the shoulder to the fingertips, 'limb, bough', etc.). Further extensions (bracchiātus 'branching' --> Proto-Romance *bracciatella- 'a little whatsit with what looks like crossed arms or branches', the ancestral form of modern pretzels) just take us farther and farther from the original adjective, both derivationally and semantically.

      It isn't quite certain that the English (and Dutch) 'merry' words really belong to the same etymological set: the vowel grade of the Dutch forms, in particular, is a problem.

  6. I was rereading Béowulf and started wondering about the etymology of Hrunting, the name of the sword that Unferþ gives to Béowulf.

    Recalling your piece, it seemed to me quite plausible that hrunt- was derived from (the precursor to) the aforementioned verb hrindan ‘to thrust, to shove’. It would then continue pre-Germanic krn̥dʰ-n- (or *ḱrn̥dʰ-n-), presumably an old participle.

    Whatever its etymology, it also underlies Old Norse hrotti, a poetic word for ‘sword’.

    - Olivier

  7. It's an ingenious idea; thanks for sharing it. The final consonant of the root could also be *t (with its Vernerian reflex generalised, as in findan). I'm surprised that Guus Kroonen doesn't discuss Hrunting/hrotti anywhere. How aboput connecting it further with *xrinþa- 'cattle' (OE hrīþer, German Rind)?

  8. Alas, I cannot take credit for it: I just checked De Vries’ ANW and according to him the connexion between Hrunting/hrotti and hrinda(n) has already been made by Kahle in 1903. He also notes hrotti in the meaning ‘Lump, Schlüngel’, besides Icelandic hrotti ‘langer Kerl’ and Old Danish runte ‘Springstange’. And so the ‘sword’-words might also be connected with Swedish skrynta ‘Rumpf’ and West-Falian schrünte ‘mageres Geschöpf’.

    Anyway, in the case of hrinda(n) I took *- dʰ- for a verbal extension, thus more correctly *- dʰh₁-. But a pre-Germanic root with *-t- might be better indeed, since it doesn’t violate any restrictions in the first place.

    Proto-Germanic *hrinþaz-, pre-Germanic *ḱrent-os- would be interesting. As it is usually understood to contain the Indo-European ‘horn’-root *ḱer-, how do you reckon it was derived? I have this notion that s-stems were deverbal in general, which would imply a pre-Germanic verb *ḱrent-e-, perhaps with a meaning ‘to thrust or shove with the horns’. But I don’t feel comfortable with that...

    - Olivier

    1. Oops, sorry for the typo, it's *xrinþaz/*xrinþiz-, of course, which is precisely why I think it looks deverbal rather than denominal. Something like *ḱrent- might account for both hrand and hrīþer, not necessarily involving the 'horn' word-family -- unless someone can explain the morphological makeup of *ḱr-ent-es-

  9. To make matters more interesting, on account of Middle Dutch, Low German runt, ront, Dutch rund and Old English hrýðer (pl. also hrúþeru), the word seems to have been an ablauting s-stem originally, i.e. nom. *hrinþaz, obl. *hrundiz-. In which case Old English continues mixed forms. But I don’t know how to further analyse pre-Germanic *ḱrent- (or *krent-), I’m afraid.

    However, for Hrunting and hrotti, if they do indeed belong with hrinda(n), perhaps they weren’t derived from an old participle, but from a lost intensive/iterative 3sg. *hruntōþi, 3pl. *hrundunanþi. When I get the chance, I shall see whether I can I find some overlooked reflexes in modern dialects.

    - Olivier

    1. I remember these apparently ablauting variants of the es-stem are discussed by Stefan Schaeffner in Das Vernersche Gesetz..., but I'm away from my Faculty library at the moment and can't check up the details.