Thy More Thy Merrier.

Frequent commenter ktschwarz made a Wordorigins post about the construction “the more the merrier,” pointing (with a sigh) to a meme that’s going around claiming that “the” here is “not the article […] it is a different word” meaning either ‘when’ or ‘then’: “‘The more, the merrier’ literally means ‘when more, then merrier.'” This is not even a little bit true; kts quotes a Twitter thread by linguist Danny Bate that begins:

‘The more, the merrier’ is an example of a comparative correlative.

The construction is formed through the pair ‘the… the…’ in Modern English – the two come from Old English ‘þȳ… þȳ…’ and are therefore relics of the instrumental, an extinct grammatical case of English.

That is to say, the ‘the’ in phrases like ‘the more, the merrier’ and the definite article come from different forms of the same word (Old English sē, sēo, þæt). The fact that the two are now identical in modern spelling and pronunciation is due to a later accident of history.

This is a nice similarity between English and Czech – the same construction also makes use of the instrumental case, which in Czech is alive and well. For example, the phrase ‘čím více tím lépe’ (‘the more, the better’) relies on the words čím and tím, both in the instrumental.

The OED has an entry the, adv. (updated June 2018), “Used with a following comparative adjective or adverb to emphasize the effect of circumstances indicated by the context”:

Etymology: Probably use as adverb of a case form (perhaps neuter instrumental) of the Germanic base of the demonstrative pronoun (see the adj., pron.², and n.¹; compare that pron.¹, adj.¹, adv., and n.), perhaps ultimately reflecting the same Germanic form as the conj.
 
Compare thy adv., thon adv.
 
In later use probably partly also a phonologically reduced form of thy adv.
It is frequently assumed that in Old English the word shows an inherited long vowel (þē) which was shortened in low stress.
 
In Old English the word, probably reflecting what was historically an instrumental form, is found beside more frequent þȳ thy adv. and þon thon adv. in the construction with a comparative (see sense 1), sometimes in the same sentence (compare quot. eOE at sense 1a). Compare also thes the at thes adv. 2.
 
With use in the correlative construction in sense 1c compare similar use of þȳ (see thy adv. 2a). (It has alternatively been suggested that the second the in this construction may perhaps reflect earlier use of the conj.)
 
With sense 2 compare thy adv. 1a. However, this causal use can be difficult to distinguish from uses of the conj. (especially the conj. 1a or the conj. 3a), and the sense of individual attestations is sometimes disputed.
 
The origin and relationship of the Old English instrumental forms þon, þȳ, and the fossilized þē̆ are uncertain and disputed; see discussion and references cited at thy adv. and pron., and compare the Germanic forms cited at that entry.

The construction under discussion is at 1.c.:

In the correlative construction the ——, the ——, denoting proportional dependence between the notions expressed by two clauses which each contain a comparative: as much (more ——), so much (more ——). Cf. thy adv. 2a.

And “thy, adv. and pron.” “For that reason, therefore” (updated June 2018) says:

Etymology: Originally (in Old English) specific uses of the neuter instrumental case of the demonstrative pronoun se, sēo, þæt (see the adj., pron.², and adj.¹). Compare thon adv., the adv.
Compare in similar use the following continental West Germanic instrumental forms in combination with a preposition (compare sense B.): Old Frisian thiu, Old Saxon thiu, Old High German diu (Middle High German diu); and forms probably instrumental in origin used with a following comparative (compare sense A. 2): Old Frisian thi, the, tho, Old High German the, thi; as well as the Gothic particle þe (see the conj. and pron.¹) attested in both uses.
 
The origin and relationship of the Old English instrumental forms þon (see thon adv. and pron.¹), þȳ, and the fossilized þē̆ (see the adv.) are uncertain and disputed; for conflicting suggestions see A. Campbell Old Eng. Gram. (1959) §709, R. M. Hogg & R. D. Fulk Gram. Old Eng. (2011) II. §5.7, D. Ringe & A. Taylor Devel. Old Eng. (2014) 389–90.

[…]

2.a. Used with a following comparative adjective or adverb to emphasize the effect of circumstances indicated by the context. Also in the correlative construction thy ——, thy ——, denoting proportional dependence between the notions expressed by two clauses which each contain a comparative. Cf. the adv. 1a, 1c.

eOE King Ælfred tr. Gregory Pastoral Care (Hatton) (1871) Pref. 5 Hie..woldon ðæt her ðy mara wisdom on londe wære ðy we ma geðeoda cuðon.
eOE King Ælfred tr. Gregory Pastoral Care (Hatton) (1871) xvii. 123 [Oft] sio wund bið ðæs ðe wierse & ðy mare, gif h[i]o bið unwærlice gewriðen.
OE Riddle 9 11 Mec seo friþe mæg fedde siþþan, oþþæt ic aweox, widdor meahte siþas asettan. Heo hæfde swæsra þy læs suna ond dohtra, þy heo swa dyde.
OE tr. Medicina de Quadrupedibus (Vitell.) xii. 268 Wifgemanan to donne, nim drige fearres sceallan, wyrc to dust[e], oððe elcor gnid on win, & drince gelome, he bið þy gearwra [?a1200 Harl. 6258B þe ȝearra] to wifþingum.

I trust that clears things up, and no one will ever again make false claims about it. (Sigh.)

Comments

  1. Isn’t it a little bit true, though? This the definitely isn’t an article synchronically, which seems like a good argument for calling it a different word regardless of the etymology.

  2. Hmm. Well, maybe a little bit true, then, depending on your definition of “a different word.” I took it to mean that the apparent similarity was coincidental, like that of ball ‘globe or spherical body’ and ball ‘dance.’

  3. J.W. Brewer says

    “The more it
    SNOWS-tiddely-pom,
    The more it
    GOES-tiddely-pom
    The more it
    GOES-tiddely-pom
    On
    Snowing.”

    It’s the Tiddely-Pom Construction, Innit?

    Also available in West Indian regional dialect: https://www.youtube.com/watch?v=FHAbj1pIT4g

  4. J.W. Brewer says

    TR’s point seems to have an implicit premise that a single “word” can only fall into a single category of words, where categories like “article” and “adverb” are treated as somehow metaphysically real (and impermeable, and mutually exclusive) rather than merely useful-on-balance. I think that premise values the supposed perfection of the classificatory scheme over the messy real-world data it purports to classify, which strikes me as bass-ackwards (which wiktionary FWIW treats as both an adjective and an adverb but not as two separate words).

  5. Fair enough as a general point, but the semantic distance between correlative the and the definite article seems a lot bigger than in the case of adjectives and homonymous adverbs or nouns and zero-derived verbs. Put otherwise, what’s the synchronic argument for saying they’re the same word?

  6. J.W. Brewer says

    I don’t think L1 Anglophones typically think of them as separate words, although perhaps that is simply evidence of a deficiency in their education, and the right perspective is that of a hypothetical alien linguistic anthropologist who knows nothing of the etymology and might hypothesize that they are unrelated homophones? Although of course most native speakers don’t know the fine details of etymology (at least not accurately!) either. But let me pose another example to TR. Is the “to” in “I feel like I’m fixin’ to die” the same “word” as the “to” in “I feel like I’m drivin’ to Tulsa” or are these two different words? The use in infinitive phrases probably falls outside the scope of our mostly-useful-most-of-the-time definition of “preposition,” but so what? Or at least “so what” is my native-speaker intuition as opposed to my alien-anthropologist intuition.

    Synchronically, one might if pressed try to explain the correlative the as an apparent instance of ellipsis/clipping. “The more [guests-or-whatever who attend the party-or-whatever], the merrier [party-or-whatever will occur than otherwise would have].” Maybe that’s completely false as a historical matter, but it’s a plausible cover story for finding the usage article-like. There’s generally some sort of noun that could take a definite article that’s somehow implied in the situation described. But whether that’s a better synchronic argument than TR’s pro-splitter argument … well, only time will show the wiser.

    Separately, calling “correlative the” an “adverb” as the OED does seems to lend some support to the claim one hears from time to time that “adverb” in English is not a coherent category but a miscellaneous/junk-drawer sort of label for things that don’t fit elsewhere.

  7. “The more [guests-or-whatever who attend the party-or-whatever], the merrier [party-or-whatever will occur than otherwise would have].”

    That’s how I have always mentally glossed such constructions, which are pretty common in English. If you ask somehow how they like their tea, they might say “the [sweeter/stronger/hotter], the better” which can similarly be thought of as an elliptical form of “the X it is, the better I like it.”

    ETA: Does this use of ‘the’ bear any relation, I wonder, to the use of ‘it’ in ‘it’s snowing’? (With or without the tiddley-pom)

  8. Synchronically, one might if pressed try to explain the correlative the as an apparent instance of ellipsis/clipping. “The more [guests-or-whatever who attend the party-or-whatever], the merrier [party-or-whatever will occur than otherwise would have].” Maybe that’s completely false as a historical matter, but it’s a plausible cover story for finding the usage article-like.

    This discussion seems to have gotten completely away from the original context. To refresh your memory, the claim was:

    “not the article […] it is a different word” meaning either ‘when’ or ‘then’: “‘The more, the merrier’ literally means ‘when more, then merrier.’”

    Are you (or is anyone) saying this is defensible, setting aside qualms about what “the same” means?

  9. Well no, that seems totally absurd. But it’s not clear (to me) anyway how one should interpret such constructions.

  10. To JWB’s question, I tend to see the two to‘s as distinct, but I’d be hard pressed to explain what hinges on that judgment or what it means.

    I don’t think the ellipsis argument holds water, though — the more guests in the more guests attend the party is pretty clearly not an NP, and in fact changing the comparative can cause a genuine article to crop up: the smarter the guests, the better the party.

  11. J.W. Brewer says

    The OP magisterially asserted that the “when more, then merrier” gloss was “not even a little bit true.” So that didn’t seem like a point worth commenting on. Although I wonder if that’s an idiosyncratic-in-context use of “when,” essentially using “when P, then Q” instead of the more standard-to-my-ear “if P, then Q”? But there would be some sort of error involved in thinking that, because someone who says (with apparent endorsement) “The more the merrier” would presumably agree with the logical proposition “if more, then merrier,” the latter proposition is the “meaning” of the first statement. Unless you get back into that morass (Lord, it’s been 36 years since I took that philosophy-of-language class …) of thinking that if sentence A and sentence B have the same truth conditions they “mean” the same thing.

    I’m not claiming the ellipsis argument withstands careful critique – it’s more a native-speaker-psychology point. That the “correlative” usage falls outside the scope of what we think definite articles do (if we consciously think about that, which is already rare), generally goes completely unnoticed by native speakers using the construction, and when the oddity of the construction (given plausible constraints on how definite articles, as such, are used) is forced to their attention, the natural response is to come up with some ad hoc theory about how the odd-once-you’re-forced-to-focus-on-it construction is some sort of plausible extended sense of the “normal” usage. My claim is thus that “ah, it must be a totally different word which by random coincidence is a homophone and homograph, as I’m aware happens from time to time in English” is NOT going to be the typical native speaker’s first response when confronted by the oddity.

  12. Slavic languages have many K/Ch – T constructions.
    Cf. Russian kak-tak, kogda-togda, kuda-tuda and čem-tem (čím – tím in Czech). One of them kak X tak i Y “as X as also Y” simply means “X as well as Y” (note English as… as…! ).
    Skolko X stolko Y means “as many Y as X”.

    You can split them in two components: (1) the correlative scheme K***-T*** (2) temporal (when then) directional, etc. semantics, including our instrumental of “what… that”.

    Back to English:

    English also has W*** and T***: where there, when then… and orthographically what that.
    English has “he said that ..”. French and Slavic have “he said what …”.
    English and French have parallel as… as… the… the… plus… plus (and even le plus sera le mieux) instead of those constructions that contrast both in form and meaning.

  13. including our instrumental of “what… that”.
    INSTR normally either has instrumental meanings (by X, with X) or it means “as X” (in capacity of X) or “like X”

    So in Slavic it is “by what more, by that better” where what stands for a variable/unknown [thing, place, time, quantity…] and that points at the same [thing, place, time, quantity].

  14. Is the construction in la plus ça change etc. equally exceptional in French?

  15. I’m more ignorant than regular LH commenters on these things, so maybe I’m missing something really big, but the “when more then merrier” feels to me truthy if not (a little bit) true, even after reading all the explanations of the etymology. I’m certainly comfortable to treat “when … then” and “if … then” interchangably in a lot of contexts, as discussed by J.W. Brewer. I’m not sure the differences in meaning are particularly consistent when present, and if anything, wouldn’t the origins in the instrumental case suggest more sense of causative rather than simply correlative relationship? It also captures the idea of a wh/th pair that seems to be relevant etymologically.

    Perhaps “where… there” would be even better, at least for people used to maths-speak. But it seems to me the problems raised with the meme’s wording are to do with the way it talks of a different word “literally” meaning when/then, not the idea that “when… then” might help explain what’s gone on in some way.

  16. PlasticPaddy says

    In ME there is an alternative construction where the first the is preceded by ay, e.g. Chaucer has
    For therby to ben esed wel he wende, And ay the ner he was, the more he brende. For ay the ner the fir the hotter is.
    In other examples the first quality is clearly a comparative form, I suppose it is possible ner = nerrer with elision.
    The ay would correspond to je in the German je…desto.

  17. In other examples the first quality is clearly a comparative form, I suppose it is possible ner = nerrer with elision.
    I’d rather assume that it’s near in its original function as the old comparative of nigh, which during the course of Middle English became the positive form and mostly replaced nigh..

  18. there is also “all the more” (all the more so in Russian is tem boleje with instrumental of a demonstrative pronoun).

    And “better THAN x”

  19. I disagree with “totally absurd”: “when more then better”, “where more there merrier” can convey the same idea.

    What the meme did: it identified the correlative scheme (which I designated as (1)) but it got the own semantics of the adverb (the element (2)) wrong. “When more then better” can still work because comparatives “more” and “merrier” already contain the necessary semantics.
    But interestingly, it used the Slavic-style correlative scheme… just as Danny Bate.

    I would say that the meme attemtps to explain the original meaning, explains it partly (but tells a wrong fact about it) and does not say anything about the formal side of the etymology.

    Danny Bate in turn does not say anything about semantics (apart of “instrumental”, but it’s a name of a form as well – and the meme’s readers may not know what it is).
    Instead Danny Bate clarifies formal etymology and notes that words are related.

  20. I disagree with “totally absurd”: “when more then better”, “where more there merrier” can convey the same idea.

    I wasn’t talking about hypothetical semantics, I was talking about the idea that that is where the expression came from. Which is absurd. It bothers me when people care more about pleasing stories than actual facts (which is pretty much all the time).

  21. I still wonder, how did it arise.

    “where the expression came from”

    @LH, the meme does not claim that “when … then… ” is where the form comes from.
    The meme: “‘The more, the merrier’ literally means ‘when more, then merrier.’”.
    This is not “totally absurd” but it is inaccurate.

  22. @drasvi: – French and Slavic have “he said what …”

    Not sure that is the case in Croatian: “rekao je da…”

  23. @zyxt, thank you!
    It was a blunder. Of coruse it is Russian, not Slavic:((((

  24. Drasvi,

    When they say “literally means, read “etymologically meant.” This is a colloquial way of talking about word etymologies in English.

    To drasvi’s point, it is an interesting usage that doesn’t yet seem well explained. Or as likely, I’m just dense.

    I already feel I’m getting into swampy ground when the expert linguist offers an example that (as far as I can tell) doesn’t use the same article or article-like pronoun to introduce the two halves of the phrase he posits as similar, the way it works in English:
    >the phrase ‘čím více tím lépe’ (‘the more, the better’) relies on the words čím and tím, both in the instrumental.

    Or maybe cim and tim are forms of the same word?

    I don’t know Czech at all, but when I plug tím in online, it gives ‘preposition meaning ‘thereby’ (already problematic, since I don’t think thereby is a preposition.) Tim may be a relative pronoun used as a preposition? But already we’re getting closer to “when / then”. The online translators seem to suggest the Czech phrase means something more like “by more, thereby better”. Which seems to make it a poor analogy for explaining why the English shouldn’t be understood as “literally” when/then. (I accept the that proposition that it’s not. I’m just trying to understand what exactly our English construction is.)

    If I’ve caught on to what the instrumental case is, I’m having trouble understanding why both sides of the sentence are in the instrumental. If I recognize modern English as using prepositions to express the instrumental, I’m still lost, since I can’t really understand a sentence in the form “with the (comparative) / with the (comparative)”.

    The closest I can come is “Live by the sword, die by the sword.” Is that a single sentence with some sort of correlative comparative? Or two sentences, whose comparison is merely understood because of their apposition? Not unlike saying “Fuscia? Pink?” implying What’s the difference?

    The OED seems as lost, littering the discussion with probablies, and offering another example which like Bate’s, is not parallel, because it uses different constructions for each phrase:
    >as much (more ——), so much (more ——)

    If the English is to be understood as article + noun, in whatever case, can someone explain why it can’t accept a verb to relate them? In normal English, to my mind, two phrases are either the two sides of an identity, or there is subject and predicate. Both “bigger is better” and “The bigger the better” work for me. But “The bigger is the better’ doesn’t work. Why? It seems like it doesn’t work because these aren’t really noun phrases in modern English.

    The OED describes them as “two clauses.” That does seems to describe how they work. But how exactly are they clauses? Is there a verb implied? “The bigger (something is), the better (it is)”? And that would still leave the question of why we consider this is one sentence, since we don’t typically join clauses without conjunctions.

    The best explanation I can see is the OED aside:
    >the second the in this construction may perhaps reflect earlier use of the conj.

    In other words, etymologically, the and the were different forms with transparently different meanings, the former being instrumental and the latter being conjunctive, which is not so far from “it literally means when / then”, though “wherefore / therefore” fits more exactly. And aren’t those pairs cognate with cim and tim in Czech? wh- / th- and c- / t-?

    To J.W. Brewer’s point, we seem to have trouble describing (or I have trouble understanding and articulating) exactly how these very basic words work – how relative pronouns transmute so easily into adverbs and conjunctions, and how to understand what they are at any given point in time and in a sentence. And the related issue of how to understand what wh- and th- are in English. Are they roots or affixes? It may be, as he says, a futile quest to try to nail these words into categories that are “metaphysically real.”

    If I’m wrong in all this, well, the harder they come, the harder they fall, one and all.

  25. I don’t understand expressions like “the more the better” as meaning just “if more then better”, but rather more specifically asserting a proportionality:

    the [measure by which it is] more [is] the [measure by which it is] better

    Adding the parentheticals would actually make the “the” a real article, but that is obviously cheating. The original instrumental construction is closer to

    by what it is more [is] by what it is better

    expressing the equality of the increments in moreness and betterness. At least this is how I always understood the Slavic construction which is using instrumentals to this day.

  26. David Marjanović says

    Is the construction in la plus ça change etc. equally exceptional in French?

    La is wrong here. The rest is exceptional in fronting plus for emphasis: “it changes more, it’s even more the same thing” > plus ça change, plus c’est la même chose.

    Not sure that is the case in Croatian: “rekao je da…”

    Polish is different again, using że.

    Or maybe cim and tim are forms of the same word?

    No. Instead…

    And aren’t those pairs cognate with cim and tim in Czech? wh- / th- and c- / t-?

    Yes: PIE *kʷ- / *t-. In Slavic, that gives *k- ~ *č- / *t-.

    And the related issue of how to understand what wh- and th- are in English. Are they roots or affixes?

    I’d say they’re parts of related roots. The last time they were actual morphemes was some point in pre-PIE.

  27. “it literally means when / then”, though “wherefore / therefore” fits more exactly. And aren’t those pairs cognate with cim and tim in Czech? wh- / th- and c- / t-?

    English wh- is indeed cognate with Czech č-/c-/k- going back to PIE interrogative root kʷ-, and similarly with th- vs. t-. But the for the whole words it seems to be more problematic. Per Wiktionary, when is from Proto-Germanic *hwaz +‎ *-n (the suffix lacks further explanation which makes this explanation a bit suspect); then is said to come from PIE accusative. So perhaps (t/w)hen descend from PIE accusatives while (t/č)ím from PIE instrumental form of the respective pronouns. But someone knowledgeable of PIE etymologies will probably correct this hunch.

    Anyway, “čím / tím” does not mean “when / then”, it means “by what / by that”.

  28. When they say “literally means, read “etymologically meant.” This is a colloquial way of talking about word etymologies in English.” – @Ryan, I wholeheartedly agree. But the meme spoke about semantics (and semantical development). And LH said that he (in turn) was not talking about semantics….

  29. i’m a little confused by the various elaborate syntaxes being proposed behind “the bigger the better”. as everyone else has said, i may also be speaking from some unexamined confusion, but it doesn’t seem to me to be more complicated than “the bigger [it is] the better [it is]”. which seems like a perfectly ordinary parallel construction with perfectly ordinary articles, that only looks strange because english usually doesn’t like to drop verbs in that way.

    i wonder whether it’s useful to think of it in relation to longer fixed phases that do the same kind of verb-dropping, but which may (given where they’ve been preserved) have originated as specifically poetic formulations:

    “the blacker the berry [is], the sweeter the juice [is]”
    which is not customarily – but would be contextually clear if it were – abbreviated to *”the blacker the sweeter”

  30. J.W. Brewer says

    @rozele: I’m not 100% sure but I suspect that on the analysis favored by TR and the OED, in “the blacker the berry” etc., the “the” before “blacker” is not a token of the “same word” as the “the” before “berry,” because (or at least correlated to the contention that) the first “the” is not functioning as an “article.” Whatever an “article” is. (I’m headed out the door momentarily and am too lazy to double-check my vague recollection that “article” may be one of the word-categories that were jettisoned in the Huddleston/Pullum classificatory schema of English words.)

    Similarly I think the claim is that the “the” in a construction like “the bigger it is” etc. is not functioning as an article, because there is no noun it goes with and articles need a noun to go with?

  31. Regarding origins of “the more the merrier”, I was struck by that word “merrier”. We take that to be figurative in some situations where it’s used… more is better, but doesn’t actually cause any more merriness. But I was thinking, maybe it used to be more literal. And checking Wiktionary and then the Online Etymology Dictionary, indeed merry did used to have a wider meaning than it does now.

    Regarding the idea of understanding it as ordinary use of “the” with words elided, that’s how I take it. (Not here talking about etymology; just how I understand the phrase.) It also brought to mind something I remember seeing years ago, “Know God, know life; no God, no life”; it seems parallel to the 2nd part of that (“no…no…”). If there’s no God, there’s no life. If there’s more (people), (the gathering is) merrier. Similar construction, but “the more, the merrier” has more elided.

  32. Giacomo Ponzetto says

    In the good old days when English grammar was taught via Latin, they need not have recognized an Old English instrumental, but they did spot the Equivalent Latin ablative. Walker (1673):

    The before a Comparative Degree, whether Adjective or Adverb, is usually put for these Particles, by how much, by so much, and made by quo, or quanto, in the first place, and (if redoubled) by eo, hoc, or tanto in the second place: as
    The higher we are, the lowlier let us behave ourselves | Quanto superiores simus; tanto nos summissius geramus Cic. I. Offic.

    Some of us have not kicked the habit of construing the comparative correlative thusly. By however much bigger, by so much better will it be.

  33. I think the claim is that the “the” in a construction like “the bigger it is” etc. is not functioning as an article, because there is no noun it goes with and articles need a noun to go with?

    Right — whether we want to call it “the same word” or not is obviously a definitional question, but the point is that this the is not a perfectly ordinary article (nor determiner if you prefer that classification, as I believe CGEL does), either in terms of syntax or semantics/pragmatics.

  34. OED also mentions

    c. As correlative conjunction: ‘hwæþer‥þe‥’, ‘þe‥þe‥’, ‘whether‥or‥’.

    c888 K. Ælfred Boeth. xxxiv. §6 Hwæþer þincð þe þonne þæt þa þincg sien, ðe ðara soðena gesælða limu, þe sio gesælð self? 971 Blickl. Hom. 97 Hwyder he gelæded sy, þe to wite, þe to wuldre. c1000 Ælfric Hom. II. 120 Ða Gregorius befran, hwæðer þæs landes folc cristen wære ðe hæðen. c1205 Lay. 16812 Do þine iwille Whaðer swa þu wult don, Þa us slan þa us an-hon. a1250 Owl & Night. 1064 Hweþer þu wilt wif þe meyde. Ibid. 1408 Sei me soþ if þu hit wost Hweþer doþ wurse fleys þe gost. 1297 R. Glouc. (Rolls) 4507 In woch half turne he nuste, þo weþer est þe west.

  35. @JWB: hmm. i take your point. but i’m still inclined to think that older poetic phrasings may be the best place to find useful clues. (and i don’t have the time – or really quite enough old & middle english – to dig right now)

  36. The fact that the function of the in “the more, the merrier” is not its usual one as an article does not mean that its two appearances on the phrase have different functions. My idiolectical intuition strongly suggests that the is functioning the same way in each location. I don’t think that’s problematic at all, if the is functioning as a conjunction. Standard subordinating conjunctions can be fronted:

    I’m happy, because you’re here
    Because you’re here, I’m happy.

    That’s not identical, obviously, in particular because “the more, the merrier” has two appearances of the. But that’s closely related to the apparent fact that the two halves seem to be on equal ground. One is not actually subordinated to the other. It feels analogous to if we could both front and repeat a coordinating conjugation, like

    * And I’m in the car, and I’ll be there in ten minutes.

  37. Maybe it’s interference from other languages, but to me it does feel like the first clause is in some sense subordinate to the second (like in the paraphrases with if etc.). Of course, if it’s a subordinate clause you’d think it would be omissible, but *The merrier by itself is ungrammatical; but then there are things like All the better, So much the worse, which look like a similar construction without the first clause. And for that matter either clause of a coordination should be omissible too. So maybe this is neither coordination nor subordination, but a secret third thing.

  38. > It feels analogous to if we could both front and repeat a coordinating conjugation

    That’s legitimate in Spanish — “o ______, o ______.” But I’m not convinced English has/had a special case of front-and-back conjunctions that only worked with pairs of comparatives.

  39. PlasticPaddy says

    @Ryan
    the nearest thing is something like “so far, so good”, where the effect is not linear but likely to collapse in to chaos with a small increment after a certain point is reached.

  40. I honestly don’t know what subordination means here. And I don’t know what they mean about “it can’t stand alone”.

    But I suppose in English the relationship is asymmetrical (“the merrier, the more” is different) and this asymmetry is expressed by means of word order.

  41. Hat’s comment about “ball” (globe) vs. “ball” (dance) being coincidentally the same reminded me of another such pair that blew my mind: “tattoo” (inked skin) and “tattoo” (drumbeat) are etymologically unrelated words that happen to be identical.

    (Apologies if the place where I learned this fact was LanguageHat…)

  42. Norwegian humle “Humulus lupulus” humle “bumblebee”

    Russian xmel’ , šmelʹ (Slavic xъmeľь… čьmeľь)

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

    FWIW, the jo in the corresponding Scandinavian construction jo flere, jo bedre is not the same as the affirmative particle that corresponds to E yes (as opposed to ja ~ ‘yea’).

    The other jo is a loan from MLG, where it is a fossilized case form of the descendant of PG *aiwaz = ‘time’. The ON æfa has lots of descendants in Danish, but that particular one developed into i or e and my guess is that that’s the reason it was replaced.

    (TIL that the epithet of Erik Ejegod means ‘ever good’ and not ‘good with owning’ as modern intuitions will make it).

  44. @Lars, what do you mean by “that particular one”?

    It must mean “one that stands in the same relationship to *aiwaz within ON as jo in MLG”, but I don’t understand how jo is related to *aiwaz:(

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

    @drasvi: The ON was ei, later Danish i and then replaced by jo.

    Anyway, the ODS tells me that the native form corresponding to MLG jo is i (or e) < ON ei, and compares that with Gothic aiw, this being a case form of aiws < *aiwaz meaning something like ‘still’. From this I infer that I should assume that MLG jo is also a descendant of (a case form of ) *aiwaz.

    This may be a bad leap of faith, but jo in MLG is one of those irritating bad links in Wiktionary: The page for jo exists so the link is blue, but nobody has put MLG in so you just land at the top of the page.

    EDIT: Hellqvist 1922 to the rescue: Sw ju = MLG jo/ju = ‘always etc.’ < OS êo/iu with OHG the same as OS (modern HG je). Nothing past OS; but at least êo has to be attested, and getting that from something like aiw is just a few easy vowel mergers and a change from falling to rising diphthong.

  46. David Marjanović says

    Interesting. *aiwaz > êo is probably regular in OS, and there’s a MHG ê of a meaning I’ve forgotten that would also be regular. It’s been reinterpreted as an adjective/adverb and given the surviving comparative eher, which means “earlier” in the north and “rather” in the south. The southern version also comes with a superlative, am ehesten (noch) “the least unlikely”.

    I’m not sure if the semantic distance to the not-quite-standard German reassurance particle eh “don’t worry, it’s in accordance with our shared hopes or sarcastic expectations” is really too large. Phonetically it fits (Bavarian [ɛ], not [e]).

    OS iu presumably makes sense as a form so unstressed it couldn’t support the long vowel…

  47. @drasvi: English has two kinds of conjunctions, called—for not necessarily good reasons—”coordinating” and “subordinating.” Huddleston and Pullum consider them to be complete different parts of speech. In any case, regardless of what they are called, they have somewhat different grammar. The key difference, as I see it, is that subordinating conjunctions can be fronted, while coordinating conjunctions cannot. I discussed this previously here.

  48. “so far, so good” actually feels to me like the only proper parallel we’ve talked about so far!
    i don’t have anything to say beyond that, but i’m glad to see another case that feels really identical in structure to me!

    * And I’m in the car, and I’ll be there in ten minutes.
    though this feels quite close also, and for me is an entirely cromulent utterance – a bit more naturally as an SMS than vocally, but still okay out loud. the two “and”s operate as a single unit, to me, to indicate a particular time-relationship of immediacy and urgency (is that a tense/mood/aspect kind of function?).

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

    @DM, DWDS has the full story (of course). TL;DR: It’s not from the nominative, but from the accusative (WGer) or dative (NGer) of the same word as Du eeuw ‘century’ (and further out ævum, αἰών, आयु, *h₂óyu). The semantics seem to have gone from life force > lifetime > long time/forever (G ewig).

    Danish still has native ej (probably another case form) as an (obsolescent) negative particle (ON ne .. ei ‘not in my life’, ‘ne … jamais’), besides ikke (ne … eitt gi ‘not a thing at all’, ‘ne … rien’). And various opaque first members of compounds, some of them borrowed from German of varying heights.

  50. David Marjanović says

    DWDS

    *galaxy brain*

    I should have thought of seit eh und je (“since ever”, but literary). Some strange sound shifts in there, but nothing that de- and restressing couldn’t take care of, I suppose.

    Also should have remembered that my grandma has ewig “eternal” in her dialect vocabulary and pronounces it with [ɛ] there – turns out that’s etymological.

    Also: the DWDS has é as a letter you can click on in the expectation you’ll get a list of entries that start with é. But instead it just leads to Ebbe “low tide”, which has neither an é nor even an [e] and isn’t French either.

  51. J.W. Brewer says

    The simplest gloss of (ancient) Greek καί is “and,” but you can use it repetitively, with the common construction καί A καί B meaning “both A and B,” or more elaborately “not only A, but also B.” You can even use it three times in a row, as in the common doxological formula “καὶ νῦν καὶ ἀεὶ καὶ εἰς τοὺς αἰῶνας τῶν αἰώνων,”* which (of course) is translated in various ways although my personal preference is “both now and ever and unto ages of ages,” with the definite articles left out for slightly better idiomaticity (IMHO) in English.

    *In Latin “et nunc, et semper, et in sæcula sæculorum,” although I don’t recall from my long-ago schooling being taught a standard “et A et B” construction in Latin, though that doesn’t mean there isn’t one.

  52. David Eddyshaw says

    Kusaal ka, which means “and” when joining clauses (“and” joining noun phrases is expressed by , homophonous with “with”) is nevertheless at least in some sense always subordinating: the following clause must lack the tonal and segmental markers of main-clause status:

    Biig la luya. “The child’s fallen down.”
    Ka biig la lu. “And the child’s fallen down.” (cf Biig la lu zug “because the child’s fallen down.”)

    This sort of thing is quite common in Oti-Volta languages, though the details tend to vary somewhat. A good many of the languages have distinct verb flexions for the two cases, instead of the relatively simple Kusaal setup with a fixed tone overlay and an invariant enclitic particle. It’s rather like always having to use the subjunctive mood after “and.”

  53. David Marjanović says

    I don’t recall from my long-ago schooling being taught a standard “et A et B” construction in Latin

    I do (“A as well as B”, “both A and B”), and it’s still common in French.

    Preposed et occurs alone, too: et in Arcadia ego.

  54. Latin et discussed in this thread (and et in Arcadia ego starting here).

  55. If anyone is curious, Russian Academic Grammar’s treatment of “subordinate”.
    In Russian.
    http://rusgram.narod.ru/2753-2764.html#2756

    Just noticed (when trying Google tranlsation to English) that Russian подчинение – a calque of “subordination” usually means “submission”. An act of submitting [yourself] rather than making someone submit.
    подчинить is “to subordinate” подчиниться “to submit, to obey”.

  56. Subordination as the opposite of insubordination.

  57. David Marjanović says

    Insubordination is now a technical term in comparative grammar, too.

  58. ktschwarz says

    Hungarian uses the adessive case (prototypically spatial, “next to, at, by”) for correlative comparisons:

    Minél nagyobb, annál jobb.
    By-what bigger, by-that better.
    “The bigger, the better.”

    Conveniently for this translation, English “by” has both spatial-proximity and comparative-extent (“shorter by two inches”) senses, making the -nél/-nál intuitive here, to me at least.

    I bet speakers of other European languages don’t flail so helplessly at analyzing this in their own languages as English speakers do, since they haven’t phonologically smushed their “by-what” word into homophony with their “the” word. Yet we have no trouble at all in using this construction, only in describing it.

  59. ktschwarz says

    Finnish and Estonian use the partitive case: Finnish mitä … sitä …, Estonian mida … seda …. I don’t have intuition for that, it does a bunch of things that are done by a bunch of different constructions in English.

Speak Your Mind

*