Stiff Drink.

Dave Wilton investigates the origin of the phrase stiff drink, something I’d never thought about. He writes:

A stiff drink is a strong, alcoholic one. The idiom is odd to present-day ears because stiff once had a sense meaning strong that isn’t used much anymore, except in the context of booze.

The adjective stiff, meaning rigid, unbending traces back to Old English stif. And the word retains that as its primary meaning through to the present day. But in the Middle English period, stiff began to be used to mean strong. For instance, the word is used in that sense to describe the physical prowess of William the Conqueror in the Metrical Chronicle of Robert of Gloucester, c.1300:

Suiþe þikke mon he was · & of grete strengþe
Gret wombede & ballede · & bote of euene lengþe
So stif mon he was in armes · in ssoldren & in lende
Þat vnneþe enimon · miȝte is bowe bende

(Such a stout man he was and of great strength
Great bellied and bald but well proportioned
So stiff a man he was in arms, in shoulders and in loins
That scarcely any man might bend his bow)

Stiff starts to be associated with alcohol by the end of the sixteenth century, at first in the phrase stiff drinkers, that is to say hard or inveterate drinkers. […] We see stiff applied to the booze itself by the end of the eighteenth century. […] And we have stiff drink of grog in Parson Weems’s 1808 edition of his biography of George Washington. […]

So that’s where we get stiff drink from. It’s just a fossilized noun phrase that uses an obsolescent sense of stiff meaning strong.

There is a tale that says stiff drink comes from the practice of transporting corpses in spirits to preserve them for later burial. Supposedly, hard drinkers would surreptitiously sneak a drink from the cask containing the stiff. The story is nonsense, mostly. Occasionally, a corpse of a wealthy or famous person who died far from home would be preserved in spirits. Horatio Nelson was so preserved following his death at Trafalgar so he could have a funeral in England. But the practice was not common. And while one cannot discount the lengths an alcoholic will go to get a drink, actually drinking the preserving spirits would be even rarer. It’s not the origin of the phrase, just a post-hoc rationalization/joke.

Gotta love those folk etymologies! Unfortunately, the OED entry hasn’t been updated (“fully,” as they say) since 1916; this sense is their “II.13 a. Of living creatures: Stout, stalwart, sturdy […] b. Of a drinker: ‘Hard’.”

I was surprised by the previous sense:

12. Mathematics. Of a differential equation: having a solution that shows completely different behaviour over widely different scales of time (or other independent variable).

1952 Curtiss & Hirschfelder in Proc. National Acad. Sci. U.S.A. 38 235 In the study of chemical kinetics, electrical circuit theory, and problems of missile guidance a type of differential equation arises which is exceedingly difficult to solve by ordinary numerical procedures. A very satisfactory method of solution of these equations is obtained by making use of a forward interpolation process… The differential equations to which this method applies are called ‘stiff’.

I may have known that in my math-major days half a century ago, but if so it’s long gone from my memory hoard.


  1. David Eddyshaw says

    Such a stout man he was and of great strength
    Great bellied and bald but well proportioned
    So stiff a man he was in arms, in shoulders and in loins
    That scarcely any man might bend his bow

    In Kusaal “tough”, of a person, is kandir, which also means “fat”, so Kusaasi conceptions about ideal male body shape seem to be shared with the mediaeval English. Stout fellows!

    On the other hand, kpi’euŋ “strong” is also “hard (to the touch)” and “dry”, so the sense “strong” is being arrived at from a different starting-point altogether. That’s the adjective that, applied to money, means “expensive”, and to drink, “alcoholic”, so that’s the “stiff” one. (“Hard” in the sense “difficult” is expressed by tɔɔg “bitter.”)

  2. The sense also seems to survive in the phrase “stiff note”, which often refers to diplomatic communications, but googling I see it may not be as common a phrase as I thought.

  3. I also was unaware of that mathematical sense of ‘stiff.’ It seems backward to me — I would have thought a stiff equation would be one whose solutions don’t vary that much over a wide range of conditions.

    In bridge (the card game), you can say ‘I had a stiff king’ — meaning a singleton king accompanied by no other cards in that suit.

    What about stiff in the sense of not paying someone?

  4. David Marjanović says

    Stiff wind has 29.7 megaghits; in Nautical German that’s steife Brise, where steif otherwise translates to “stiff”.

    Calculating a stiff equation seems to be like sailing against a stiff wind…

  5. @languagehat: The terminology of stiff differential equations is not, in my experience, well known except among people who do a lot of computational science and engineering. As the stiffness is generally a feature of nonlinear second- (and higher-) order differential equations, it typically does not get mentioned in undergraduate mathematics curricula. As Wikipedia notes, “It has proven difficult to formulate a precise definition of stiffness,” but a key element is that naive integration of a stiff ordinary differential equation can lead to unstable and inaccurate results, because the equation allows for a lot of variation in the solutions of short time scales. Normally, what you actually want to know is the behavior over longer times scales than the rapid variations. There are specific numerical integration algorithms that have been developed to deal with stiff ODEs. The most elementary techniques for solving them accurately involve using algorithms that implicitly incorporate the evaluation of higher derivatives into the solution extrapolation process. Perhaps more useful are techniques based on recognizing the characteristic stiff behavior, and readjusting the integration time steps accordingly—making the time intervals short enough so that the very-short-time variations in the solution don’t skew the results.

    I wasn’t familiar with this terminology until I was doing numerical work for some astronomers in graduate school. I was working on a very meta computational project,* and my boss and I had a disagreement about whether it would be important (or even useful) to use an integration algorithm that could deal with a stiff ODE. He thought that the ODE we were solving could not possibly be stiff, but I pointed out that the usual heuristics for determining that were not necessarily valid, since the equation we were dealing with was not actually an ordinary differential equation of the usual type; it was actually a functional differential equation, with the second derivative depending not just on position and velocity, but also on the history of those quantities.**

    @David L: I think the term stiff in the context of differential equations arose by analogy to a coupling to a stiff spring—that is, one with a high spring constant and thus a rapid temporal response. The stiffness thus leads to high-frequency oscillations that dominate the short-time behavior and make it difficult to get at the longer-time behavior that is probably of greater interest. I am not sure whether I read this explanation somewhere, or if I intuited it myself when I was studying the subject. However, according to the Wikipedia article I linked, that is the generally agreed explanation for the “stiff” terminology (although it does seem to be unclear who actually introduced the term).

    For the other sense of stiff you mention (“not paying someone”), I would guess that it is instead related to a different line of senses, the ones previously discussed a bit here.

    @David Marjanović: In weather terminology, a stiff wind is one step stronger than a strong wind. Characteristically, in a strong wind, it may be difficult to keep a hold on objects, while in a stiff wind it is actually difficult to walk against the strength of the gale.

    * I was writing a program, which would take data from the user about how we wanted to operate the experimental device that was ultimately our project’s object of study. That program, which I wrote by hand, would then create another program, which could be run to operate the apparatus. However, most of the functioning of the second programs was itself actually just programming the digital current loop controller than actually supplied the voltages to move the system. This seemed, as I said, very meta to me at the time (summer 2000), although I’m sure current Web application programming goes many levels deeper. It was also unnecessarily awkward for the year 2000, since the “fancy” digital controller still only supported internal programs that could the fit into an array of 1024 16-bit integers. So a lot of the function of the second-level program was uploading new chunks of the full program into the controller, once it had executed a sufficiently large fraction of what of the previously uploaded program.

    ** In Kieth R. Symon’s Mechanics (one of the most exhaustive undergraduate-level texts on classical mechanics; a colleague called it “a junior Goldstein“), in his discussion of one-dimensional motion in the first chapter, it states that the most general force law for a one dimensional system has the form F(x,v,t), before dealing with specific methods for solving the less general cases that usually appear. Although I generally like the book, I feel that it is important that his statement is wrong, since the most general source law can depend not just on the position and velocities of the particle right now, but also their entire time histories up to this point. (In this case, we have, rather than an ordinary differential equation, a functional differential equation.)

    Of course, the ultimate laws of physics (so far as we know) do say that the state of the universe in the future only depends on its state now. This is known as the “Markov property,” and it is most commonly discussed in context not of deterministic systems but of stochastic processes, in which cases deviations from Markov behavior are more likely to be important. However, while the whole universe really is Markovian, subsystems (when considered on their own) need not be. In fact, almost any nontrivial approximation one imposes on a problem runs the risk of destroying the Markov property for the reduced/approximated problem. For the kind of apparatus we were working with, featuring active external feedback that was being managed by those programs I was producing, the Markov property could not even be expected to hold approximately (in general, that is; there were significant period during which the behavior of the system-plus-controller was indeed Markovian, but the transitions between these periods were not Markovian).

  6. The “stiff drink of grog” has made it to German as “ein steifer Grog” and strong intensity “steif” is also used with Wind, Brise und Frost.

  7. David Marjanović says

    Where is that geographically? I’ve never seen it with Grog, Wind or Frost (though Wind at least would not be surprising).

  8. Christopher Culver says

    The sense “strong” survives, I immediately thought, in the phrase “stiff burdoun”. I was absolutely sure this was from Joyce’s Ulysses (referring there to Boylan), but a web search suggests that I remembered it from Chaucer, and so obviously it isn’t a survival in Modern English.

  9. “Stiff” in another sense of strong is also used in phrases like “a stiff talking to”, meaning a scolding.

  10. This immediately put me in mind of Beryl Bainbridge’s beautiful and disturbing novel, The Bottle Factory Outing. I’ll leave it at that.

  11. Trond Engen says

    A wind of Beaufort class 7 (28-33 knots) is stiv kuling in Norw. and Danish, styv kuling in Swedish. Like much maritime vocabulary this could likely be Dutch,

    Stiv drink also exists, but I think that’s extracted from recent half-translations like e.g. stiv martini.

  12. “Steifer Grog” is from the north of Germany and you can even find recipes online.
    I suppose that “steifer Frost” is fairly old-fashioned but it’s around as confirmed by an online search. I don’t think it’s regional.
    Steifer Wind is 7 Bft as mentioned by others so it seems to be an international term.

  13. “Stiff drink” is not odd to my present-day ears. In Merriam Webster’s list of senses of stiff it is 4b, and to my mind rhymes into senses 4a, 6, 7 and 8. Also relevant are sense 1d “drunk” and the noun “stiffener” for a drink used to settle or steel one’s nerves.

  14. Also attested in OED, and still encountered often enough in southern climes:

    g. Unlucky. Australian and New Zealand slang.

    1930 Bulletin (Sydney) 3 Dec. 22/3 ‘’Struth!’ gasped Chips. ‘If we’re not stiff! Nothing doing for two days, Tommy.’

  15. There’s also “pay a stiff price”.

    I wonder whether a stiff wind means one that’s hard to walk into, like a stiff door is hard to open. Could you have a stiff wind behind you?

  16. The bob-and-wheel of stanza 2 of Sir Gawain and the Green Knight is:

    If ȝe wyl lysten þis laye bot on littel quile,
    I schal telle hit as-tit, as I in toun herde,
    with tonge,
    As hit is stad and stoken
    In stori stif and stronge,
    With lel letteres loken,
    In londe so hatz ben longe.

    Tolkien translates it thus:

    If you will listen to this lay but a little while now,
    I will tell it at once as in town I have heard
    it told,
    as it is fixed and fettered
    in story brave and bold,
    thus linked and truly lettered,
    as was loved in this land of old.

    The references in the wheel (last four lines) are to the alliterative meter.

  17. @JC thus stif > brave as in stiff upper lip?

  18. January First-of-May says

    Gret wombede & ballede · & bote of euene lengþe

    I wonder if there’s any other instances of womb referring to a man’s belly. I don’t recall offhand what the etymology is…

    As for ballede, I’m sure there’s a good semantic reason for translating it as “bald” rather than “(great-)balled”, but it still looks a bit hilarious… could it have been referring to some different, less obscene, body part? I can’t figure out how “bald” would fit the context either, it’s hardly a sign of a strong man – though perhaps the Middle English opinion differed.

  19. PlasticPaddy says

    line 8008-8009 (Gret-wombed… is from line 7731)
    Þer ne ssolde no mete ne drinke · bote it ouer dere ·*.
    Come wiþinne is wombe · ne cloþ ouer is suere

  20. any other instances of womb referring to a man’s belly

    Many, up to Early Modern English. For example:

    Sir Gawain & Green Knight His wombe & his wast were worthily smale.

    Shakespeare, Henry IV, Pt. 2 Falstaff: I have a whole school of tongues in this belly of mine, and not a tongue of them all speaks any other word but my name. An I had but a belly of any indifference, I were simply the most active fellow in Europe: my womb, my womb, my womb, undoes me.

    This is the oldest sense in English. Some of the Germanic cognates still have the general sense of ‘belly, paunch, abdomen’, some have specialized to the stomach of an animal or particularly a ruminant. Previously at Language Hat: 2020, 2022. No known etymology beyond Germanic.

  21. David Eddyshaw says

    I can’t figure out how “bald” would fit the context either, it’s hardly a sign of a strong man

    Well, eunuchs don’t go bald.

  22. PlasticPaddy says

    I don’t know how far back this goes, but fighters of the brawling sort are known to shave their heads in order to (1) present a more fearsome appearance, e.g., when one’s hair is like Boris Johnson’s;
    (2) prevent an opponent from seizing one’s hair and holding or pulling, so as to inflict pain, cause one to lose one’s balance, ram one’s head against a hard surface, etc. For female brawlers there would seem to be unwritten rules, obviating the need for head-shaving.

  23. Trond Engen says

    Suiþe þikke mon he was · & of grete strengþe
    Gret wombede & ballede · & bote of euene lengþe
    So stif mon he was in armes · in ssoldren & in lende
    Þat vnneþe enimon · miȝte is bowe bende

    J1M: I wonder if there’s any other instances of womb referring to a man’s belly. I don’t recall offhand what the etymology is…

    ktschwarz has answered. I’ll just add that Norw. vom used for a man always means a big belly.

    As for ballede, I’m sure there’s a good semantic reason for translating it as “bald” rather than “(great-)balled”, but it still looks a bit hilarious… could it have been referring to some different, less obscene, body part? I can’t figure out how “bald” would fit the context either, it’s hardly a sign of a strong man – though perhaps the Middle English opinion differed.

    I immediatly thought “buttocks”, which fits with being a þikke man. The meaning here may be more “broad” than “fat”, though, so a man with a huge torso and strong limbs.

  24. Christopher Culver says

    womb as generic ‘belly’ for any sex in Middle English is also found early in Piers Plowman: “hastest the homward for hunger of thy wombe.”

  25. Lars Mathiesen (he/him/his) says

    Danish vom is indeed ‘paunch’ (notionally more common in men, but I think that’s more a prejudice about physiology than a matter of semantics) as well as ‘rumen’ (specifically, as opposed to the other stomachs of the cow).

  26. Proto-Germanic *wambō has quite a list of descendants, including Medieval Latin gambesōnem.

  27. David Eddyshaw says

    Scots wame means “belly.”

    [ninja’d by Hat]

  28. Well: I. n. 1. As in Eng., the womb, uterus; 2. In various senses of Eng. belly; 3. Used transf. as the seat of the passions or of the thoughts: the heart, mind, head. Cf. the similar semantic development of Gr. φρενες; 4. Specif., of tripe or viscera used as food; 5. In transf. senses: (1) a hollow, a cavity. (2) a sufficiency of space, room. (3) the bottom row of peats in a bank (Cai. 1973).

    Some nice quotes under 3:

    Sc. 1817 Scott Rob Roy vi.:
    “Why, Andrew, you know all the secrets of this family.” “If I ken them I can keep them,” said Andrew; “they winna work in my wame like barm in a barrel.”
    Knr. 1886 H. Haliburton Horace 31:
    There’s Watty wi’ the budget in his wime.
    Ayr. 1887 J. Service Dr Duguid 133:
    I whummled Tam’s case through my wame ae nicht with a hue of toddy.
    Ags. 1927 L. Spence Weirds & Vanities 1:
    Wi’ Vergil in my loof Troy warked sae greatly in my wame.

  29. Where is that geographically? I’ve never seen it with Grog, Wind or Frost (though Wind at least would not be surprising
    I haven’t seen steif used with Frost, but with Grog and Wind it’s not unusual.

  30. I was thinking of the Yeats lines

    “Though stiff to strike a bargain,
    Like an old Jew man,
    Her bargain struck we laughed and talked
    And emptied many a can;”

    Leaving aside any criticisms of the ethnoreligious-etc. specificity of the simile, I’m trying to figure out exactly what sense of “stiff” is involved. You can find in the google books corpus (typically but not exclusively in older sources from Yeats’ lifetime or earlier) the phrase “stiff bargain,” which I don’t specifically recall encountering in my own lifetime. Or is the stiffness less about the substance of the bargain and more about Mrs. Moore’s bargaining style, maybe sense 2a or 2b in the Merriam-Webster list linked above?

  31. From the (1916 vintage) OED entry:

    8. figurative.
     a. Inflexible of purpose, steadfast, resolute, firm, constant.

    b. In an unfavourable sense: obstinate, stubborn; not amenable to reason. Now rare.
    1681 J. Dryden Absalom & Achitophel 17 Stiff in Opinions, always in the wrong.
    1838 T. C. Haliburton Clockmaker 2nd Ser. vii. 104 Considerable stiff folks, in their way them quakers—you can’t no more move ’em than a church steeple.

    10. Of price, charges, rates, etc.: unyielding, firm; having an upward tendency. Hence of a commodity or the dealers in it. Cf. A. 20.
    1883 Manch. Examiner 14 Dec. 4/1 For three month’s bills the terms were firm at 2 5/8 per cent, but for January paper the rate was stiffer.
    1886 R. Holland Gloss. Words County of Chester (at cited word) A butcher will tell you ‘You’re very stiff this morning’ if you will not come down at all in the price of a beast.
    1888 Daily News 5 Nov. 7/2 Buyers..find sellers stiff.
    1893 Daily News 14 July 3/7 The latest reports from London show that merinos are a little stiffer.

    Those seem to be the possibly relevant senses.

  32. Kate Bunting says

    The first time I came across a reference to a stiff drink, as a child, I understood it to mean ‘partially set’, as though it had gelatine in it!

  33. Though stiff to strike a bargain

    Whatever is the exact dictionary rubric, stiff here obviously means that Mrs. Moore were reluctant to bargain down. Contrast with a common phrase drive a stiff bargain where it means bargaining down a lot.

  34. I only know drive a hard bargain with that sense.

  35. Looking at google hits I see that common was an exaggeration. But it does exist.

  36. J.W. Brewer says

    The OED’s old sense 10 as quoted by hat would imply that (as in the Yeats poem) a “stiff” bargain is one with a price comparatively advantageous to the seller (literal or metaphorical) and disadvantageous to the buyer (ditto), whereas D.O.’s claim about “drive a stiff bargain” is that the point of view is the other way round, with the price being comparatively advantageous to the buyer and disadvantageous to the seller. Is it possible that the verb DRIVE in “drive an ADJ bargain” forces or at least facilitates adoption of one POV rather than the other because of some implicature about who (as between buyer and seller) the “driver” is? There are I guess also perhaps different default expectations in different sorts of contexts as to whether buyer or seller will be the first to name a specific price which the ultimate negotiated price is then either fairly close to or fairly distant from.

  37. J.W. Brewer says

    Moving on from Yeats I’m now thinking of the more recent wordsmith Lou Reed and the lines

    But you know it could be a hassle
    Trying to explain myself to a police officer
    About how it was your old lady got herself stiffed

    The noun “stiff” for corpse is perfectly standard, if informal, but the passive construction “to get stiffed” seems less so (and there is AFAIK no corresponding transitive verb “to stiff”) yet is in context perfectly transparent semantically.

  38. I would say “perfectly transparent” is an exaggeration. That would certainly occur to me as a possible reading, but the complete unfamiliarity of the usage would lead me to wonder if that’s what it actually meant. This is why it’s not a good idea to follow the usage rules of Humpty Dumpty.

  39. Perhaps it’s less “perfectly transparent” quoted in isolation. In context, the possibility-to-probability (although I guess not 100% certainty) that the lady in question has died by overdose has already been introduced into the discourse two verses previously, which helps guide the listener smoothly to the correct parse.

    Although of course weird variants on conventional use (whose semantics make sense when you stop and think about them) were a common stock-in-trade for poets in the 20th century and probably earlier.

  40. True, true. My inner curmudgeon made an unexpected foray into the outer world.

Speak Your Mind