Since we’re on the topic of myths, here’s Stan Carey’s delightful A to Z of English usage myths:
English usage lore is full of myths and hobgoblins. […] Huge effort is wasted on such trivialities. So, as a quick exercise in myth-busting (and amusing myself), I posted an A to Z of English usage myths on Twitter last week. Reactions were mostly positive, but some items inevitably proved contentious, as we’ll see.
You can click through on this initial tweet for the full A–Z plus supplements on Twitter, or you can read the lightly edited version below, followed by extra notes and quotes now that the 140-character limit doesn’t apply.
It starts with
A is for ALTERNATIVE. Peevers say you can’t have more than two alternatives, because Latin. This is the etymological fallacy.
… and goes all the way to
Z is for ZOOM. Theodore Bernstein said that this term, being from aviation, should only mean ‘upward mobility’. English went ¯\_(ツ)_/¯
He says “A few readers were happily on board until they reached a particular bugbear”; I’m happy to say I was happily on board throughout (though a few items raised a nostalgic smile for my own long-ago peever past, when I complained about things like the “misuse” of transpire). I’m also happy to report that there’s no actual peevery in his comment section, though there are a few cavils. And I agree with John that “learning to say ‘I personally object to what that person is doing, but I accept there’s nothing objectively wrong with it, and no, nobody Ought To Make A Law Against It’ would for many people be a useful and necessary lesson in tolerance outside of the matter of linguistics.”
One curious and seemingly contradictory phenomenon I’ve noticed among linguists is that they tend to oppose prescriptivism, while at the same time denouncing the widely popular usage of linguistic terms (such as “passive voice” or “grammar”) that runs contrary to their precise technical meaning.
Has any linguist ever tried to reconcile these positions? (Or argued that if other forms of prescriptivism are wrong when they oppose established usage, then it makes no sense to oppose, for example, the well-established usage of “passive voice” for any way of speaking that tries to be evasive about agency and responsibility?)
Vladimir: Personally, I think that if there are folk definitions for linguistic terms, I don’t mind people using them, but if supposed language mavens are misusing terms when they try to appear learneder-than-thou, they deserve the abuse they get.
Put another way, I am all for Herman Melville calling a whale a fish, but a whale expert wannabe has to know better.
Objecting to the misuse of technical terms is not prescriptivism as such. The problem arises when a term has both a popular and a technical sense, and the speaker tries to claim the authority that goes with technical claims while using the terms in a popular sense. This is particularly frequent in linguistics because people, even educated people, are more ignorant of it than almost any other discipline. As Mark Liberman says:
In discussing technical terminology, it also matters whether the technical word for something was actually coined for that purpose, or whether it was a preexisting word pressed into a newer (or more precise) meaning. I am much more forgiving of people using “energy” vaguely than “entropy.”
Computer people actually do get vexed at people who misuse our technical terms like “parameter” or “glitch”. But that doesn’t stop us from stealing other people’s words like “cookie” or “boilerplate”.
My unfavorite computer terminology misuse is when I’m summoned to help someone with their computer having “crashed”, and finding out that they can’t open a website because they mistyped something. Hasn’t happened to me for a while though.
Transpirare comes from trans + spīrō. The latter is, naturally, the street name for spirolactone, frequently used as a testosterone blocker by transgender women in certain countries (while the former is an umbrella term for transsexual, transgender, gender-fluid etc.) So clearly the proper sense of “transpire” is sense #4 from Wiktionary: “To become known; to escape from secrecy”, specifically a transgender coming-out event (though it’s easily extensible to other kinds of coming-out).
He says to “look up” irregardless, as if something being in a dictionary is any indication of whether the word is actually used.
(not that I’m objecting specifically to irregardless being a word)
Here’s a tip for using dictionaries: Read past line 1.
My phone burst into flames as it displayed this sentence.
@Vladimir the widely popular usage of linguistic terms (such as “passive voice” …
“passive voice” as in the two words together has always been a technicial grammar/linguistics piece of terminology; and applies to Latin (and some other inflected languages), but not to English — where the passive is marked by word order and constituents of a clause [Pullum passim].
So the popular usage just means ‘passive’/weak-and-I’m-trying-to-blind-you-with-my-superior-intellect-by-using-some-fancy-extra-word. If you look at those popular usages, there’s no coherent language feature(s) they denote.
By all means criticise corporate blandness and blame-shifting, or public figures’ evasion of responsibility. That’s criticising their (lack of) behaviour. It seems to me that criticising their language is turning it into a relatively trivial point.
Computer people actually do get vexed at people who misuse our technical terms like “parameter” or “glitch”
But ‘parameter,’ I believe, is a term of mathematical origin, and ‘glitch’ (according to various online references) was originally used by engineers, possibly in early NASA projects, to denote a damaging voltage spike.
Yes, that’s another problem: people in specialized fields who naturally assume that the terms they use are “theirs,” not realizing that they themselves may have (mis)appropriated them from other specialized fields. It’s a round robin of theft and misuse!
I agree with Vladimir on the word “grammar”, it’s pretty prescriptivist and peevish to insist that it only holds the technical meaning from linguistics/mathematics/computer science. The word is widely used to mean “rules for language” (not just syntax, but also spelling, punctuation, semantics, etc.), and that’s fine in non-formal contexts. It can retain it’s formal technical meaning while still meaning something different to regular people.
“Passive voice”, however, is a highly technical term, and people who use it intends the technical meaning, it’s just that they’ve misunderstood the technical meaning. The “energy”/”entropy” example from physics that @Brett suggests is a good analogy.
David L: I was curious, so I looked it up: the two earliest references for “glitch” in the OED are both from 1962. One is a book by John Glenn called “Into Orbit” (which is where the NASA connection comes from), the other is from the “Modern Dictionary of Electronics”, where it’s defined as a “low-frequency interference in a television picture. It is seen as a narrow bar moving vertically.”. I’m guessing it didn’t originate at NASA, it was just a general informal engineering term around that time.
It’s a round robin of theft and misuse!
Northrop Frye: “I do not claim that I am correctly interpreting Coleridge’s term, but the necessity of being a terminological buccaneer should be clear enough by now.”
But it does piss me off when poli sci people use parameter to mean ‘limit’.
@John Cowan – isn’t it really that the differential between what people think they know about language, and what trained linguists think they know about language, is greater than for any other subject matter, vs the discipline normally associated with its scientific investigation? Liberman’s point isn’t only or mainly about education level–most people don’t know very much about any particular discipline–it’s about the level of assumption lay people make about the subject matter, and the freedom they feel to propound upon it. In some ways, this can be seen as a natural and not-so-terrible-after-all consequence of the descriptivist doctrine that almost all speakers are experts in their own language (so defined), and also a good sign of the near pervasive interest in the subject matter of language. LH’s welcome admission of a peevish pre-self I think must ring true for a lot of us who love language.
Peevers say you can’t have more than two alternatives, because Latin. This is the etymological fallacy.
Yet for some reason I never hear of anyone claiming that a “combination” can only consist of two elements, even though the Latin there is, if anything, even more transparent. (Hint: it’s the same root as “binary”.)
And on a barely related tangent – the Russian word for “combination” is сочетание, of Old Church Slavonic origin, and, at least on the surface, exactly parallel to the Latin term; weirdly, this parallel had apparently been missed by Piotr Gąsiorowski in his otherwise excellent etymology of “four”, where he apparently claimed that there is no South Slavic evidence for чета meaning “pair” (as opposed to “troop”).
(To quote a comment on one of the respective pages: “The meaning ‘troop’ can be equated with ‘multitude’ and derived from a collectyive meaning ‘multitude pairs’ or, while we are at it, ‘multitude of anything at all’.”
I briefly wondered whether it makes any sense for said meaning to be back-formed from сочетание – though, best I can tell, it doesn’t quite fit the chronology anyway.)
I think prescriptivism has its place.
I object to slurs, unless people have whichever-word privileges, and I don’t see common usage as a legit defense– rather a common problem.
Some of these slurs target specific groups, and some of these slurs are associated with abuses such as forced sterilization. If people use these slurs as their favorite go-to insults, I think it reinforces prejudice, discrimination, and often violence against these groups. For example, I think calling other people “m****(s)” or “i******(s)” can reinforce the prejudice, discrimination, and endangerment against intellectually and/or neurologically disabled people.
… That that puts me at odds with most newspapers.
But that’s not the prescriptivism we’re talking about; in this context, “prescriptivism” doesn’t mean “telling people they shouldn’t do or say something or anything,” it means “telling people the language they use is wrong on the basis of ideas of correctness they’ve picked up god knows where and cling to despite all evidence.”
AntC:
I think you’re mistaken about this. As far as I can discover, a) “passive voice” is a perfectly cromulent linguistic term with respect to English¹, and b) Pullum has never claimed otherwise³.
[1] It occurs, for example, as a section heading in chapter 16 of The Cambridge Grammar of the English Language by Rodney Huddleston and Geoffrey K. Pullum².
[2] Admittedly not a chapter authored by Pullum.
[3] True, he seems not to be especially fond of the term “voice”, at least when addressing a general audience; in this Language Log post from 2011 he notes that he’s avoiding the term, in that essay, because it “mainly confuses people”. He does not, however, make any suggestion that it is not the standard term for the grammatical category to which passive and active belong, or that it is less applicable to English than to other languages, and if he has made such claims elsewhere I did not find them. In a 2014 paper, “Fear and Loathing of the English Passive”, he uses it himself: “It is not verbs that exhibit passive voice in English, but larger units.”
Incidentally, anyone who is interested in what Pullum has to say about the English passive might enjoy this series of six short videos on youtube. The first half, at least, I found quite a nice presentation of the grammatical facts (the second half focuses on examples of bad usage advice which attack the passive, often without being able to identify one).
There was something in a comment that I was trying to post that so irked the demiurge of WordPress that it wouldn’t let me post and blocked me from editing a placeholder comment when I tried to alter it post-post. That placeholder comment is now gone…
It would be nice if the software would explain itself on occasion.
I know, it’s very frustrating.
@Tim May, no. In ‘Fear and Loathing’, Pullum describes “passive clauses”, “passive construction” or just “passives”.
The collocation “passive voice” occurs only when he’s quoting others. And that one quote you found I think is again mention not use — there should be scare quotes round the term. I read it as the ‘not’ having wider scope: it is not that verbs exhibit the passive in English, but larger units. If it were verbs, we could talk of the verb exhibiting passive voice.
Perhaps you’d like to list some verbs that exhibit passive marking? Be careful to exhibit the marking that differentiates them from plain participles. Be careful to show which auxiliary exhibits the passive, and what differentiates its use as passive marker from other uses of that auxiliary. Be careful to explain how we can tell a verb is passive when there’s no auxiliary, as in an embedded/adjunct attribution.
I think you’ll find as Pullum describes: the telling symptom is a ‘by …’ PP, or that we can insert a ‘by …’ PP to stipulate the agent. The form of the verb alone tells us nothing.
What’s the basis of your claim (if I’m understanding you correctly) that “passive voice” can only refer to a verb and its bound morphology? Voice involves, by definition, a rearrangement of a verb’s argument structure. It’s a syntactic(-semantic… and there’s often information-structure involved too, but the relationship between the verb and its arguments is the core of it) phenomenon that can involve case marking on NPs, bound morphology on verbs, use of auxiliary verbs, etc.—depending on the language in question. I can see objecting to “passive verb” to refer to English, but I don’t see the necessary connection between “voice” and “bound verbal morphology”.
I was reading the link with an open mind until I got to “‘literal meaning’ is literally a contradiction in terms. The peeve is (figuratively) hoist with its own petard.”
Anyone who writes “hoist with” instead of “hoist by” clearly doesn’t know what a petard is. So I stopped there.
Not that I have any pet peeves or anything.
Man, I hate to do this to you, but here’s Shakespeare:
@Bloix: I only know what a “petard” was because my brother-in-law starred in a college production of Stoppard’s The Real Thing. In the play, the expression (and the fact that nobody knows what a petard is) gets discussed. My brother-in-law, curiosity piqued, looked it up.
Bloix: Anyone who writes “hoist with” instead of “hoist by” clearly doesn’t know what a petard is.
i am not entirely sure myself (although I understand the whole phrase), but I don’t remember coming across either of these two combinations. “Hoist on” is the only one I know.
In Russian, the word петарда is still extant, but it means a piece of fireworks – not really something one can get hoist with, on, or by.
I don’t recall what English petard means either (I saw it mentioned somewhere, but forgot).
The English word is a borrowing from French le pétard, a small piece of fireworks used as a noisemaker. The word is derived from the verb péter ‘to fart, to burst’.
… Or used to blast a gate open.
@ə de vivre, the peevers who rail against “passive voice” generally don’t know the passive from a hole in the ground. So for them to add “voice” in an attempt to blind with science is denoting nothing. (They are not talking about Latin or Greek; but given the vintage of the typical peever, they may have half-remembered something from school days — or perhaps their English teacher or usage maven had.)
If you want to talk about a verb’s argument structure, I suggest you’re talking about clauses — even for languages with bound verbal morphology and case marking: ablative is not a reliable indicator of a passive clause (and a clause can be passive with no ablatives in sight).
Most dictionaries I checked do define the grammatical sense of “voice” as “A particular way of inflecting or conjugating verbs, or a particular form of a verb, “[wiktionary]. But of course they must record usage. So if the preponderance of peevers when talking about a language with no verbal morphology for the passive use “passive voice” to mean something about evasiveness of agency in sentence structure, that’s what it’ll come to mean.
For some reason I can’t open Stan’s blog.
As I mentioned elsewhere, “linguist” is one of those words like “energy”. It was “linguists” who redefined the term and take umbrage when people use it in the sense that it used to have before they appropriated it.
So when “linguists” get upset that they are NOT a person who knows many languages, they are actually being prescriptivist.
True, but it’s very difficult for any of us to be cold-bloodedly objective about what’s central to our lives. We are, after all, all human.
I don’t know of anyone who actually uses “linguist” to mean a person who knows many languages. I only hear that with people trying to figure out what a linguist does. Of course that’s the original meaning of the word, but I’d say that usage is archaic by now.
You’re welcome to follow your own personal usage, all I’m saying is that linguists have used “passive voice” to mean something beyond bound verbal morphology. Unless you’re claiming that voice is in general an untenable category altogether, which is something I’d be willing to entertain, but that doesn’t change usage in general. And I’m not sure what you’re talking about with ablatives and argument structure wrt the passive, of course they’re not necessarily related…
@Y: “Linguist” meaning “polyglot” is perfectly ordinary, and I have encountered it numerous times. The United States Army is always trying to hire more “linguists,” for example.
As I recall, it is not actually clear which meaning of “linguist” is older, although both are certainly venerable.
linguists have used “passive voice” to mean something beyond bound verbal morphology.
So now we’re no longer concerned with peevers(?)
Have they? Pullum doesn’t (when talking about English). Are those “linguists” as in speakers of many languages, who’ve been taught those languages by somebody who applies grammatical terminology from Latin? Or “linguists” as in studiers of language-in-general who are careful to analyse each language in its own terms, rather than the cargo cult of treating all languages as decayed Latin (or decayed PIE)?
Unless you’re claiming that voice is in general an untenable category altogether,
I think it’s untenable for describing English (which is all the peevers are talking about). I think active/passive is a distinction applicable to many languages (including English). Old-school grammarians payed very little attention to word order, as opposed to morphology (because Latin/Greek supposedly have ‘free’ word order). So they associated active/passive(/deponent/reflexive) to verbal morphology, and needed an overarching category.
If you want an overarching category for active/passive in English — and other languages with marking through constituent order or particles rather than verbal morphology — I suggest it’s misleading to use “voice”: rather “clause” or “construction”.
I believe that “linguist” in the linguists’ sense comes from the term “linguistics”, which was originally from linguistique, the term used by Saussure. It is the newer sense of the term.
I forgot about the army usage. In common speech, though, I’ve never heard it used as such.
The first known use of the word linguist was to describe a mockingbird, or something like it.
AntC: Forgive me. In trying to be concise, it seems I have been less than clear.
My understanding of your position, in the passage which I quoted in my previous post, might be summarized by the following syllogism:
A) In English, the passive is not indicated through inflection on the verb but through syntactic constructions;
B) “Passive voice”, in its standard use as a technical term within linguistics, can only correctly be applied to languages in which the passive is marked on the verb;
C) Therefore, the term “passive voice” in its usual linguistic sense cannot be correctly used in reference to English.
(Would you consider that a fair restatement of what you said? You might perhaps draw the distinctions in A and B somewhat differently, but the precise details aren’t actually all that important to my point here.)
When I stated that you were mistaken in believing C, I seem to have given you the impression that I denied A. But that wasn’t my intention at all; what I was denying was B. I am not disagreeing with you about the nature of English grammar but merely about the meaning of the term “passive voice” in linguistics.
Now that I have said this I hope you will understand why I have no intention of listing any verbs that exhibit passive marking—I have made no claims regarding their existence and they are irrelevant to anything I have argued.
As evidence that B and C are false, here are some quotes I prepared earlier. In Trask’s A Dictionary of Grammatical Terms in Linguistics (1993), “voice” and “passive” are defined like this:
In Quirk, Greenbaum, Leech & Svartvik’s A Comprehensive Grammar of the English Language (1985), section 3.63-78 (p. 159) is headed “Active and passive voice” and begins:
In Huddleston & Pullum’s Cambridge Grammar of the English Language (2002), section 10 of Chapter 16 (p. 1427) is entitled “Passive voice“:
Incidentally, here’s a 2006 Language Log post in which Pullum uses the term “passive voice” and is not echoing anyone else’s usage, with or without invisible scare quotes.
It’s still current in the U.S. military: a linguist is either a translator or an interpreter, and cryptologic linguists are those who decode foreign communications after the cryptanalysts have worked on them.
Thanks for being clear @Tim, yes I hold C).
Trask exhibits typical confusion: “active and passive constructions”; “passive … 1. A construction …”. Then mixes up syntax with semantics: “participant roles”, “expresses an agent”, “expresses a patient”. No: if you want a syntactic characterisation (of English), talk only about word order/constituent order; talk about the NP that appears before the verb in an active construction cp. appearing in a ‘by …’ PP for the corresponding passive construction.
Q, G, L & S are clearly talking about verb morphology, and that’s not adequate to characterise the English passive.
The 2002 CGEL entry is not written by Pullum, as you noted. I guess wrt his 2006 piece (and the 2011 avoidance of “voice”) he has to mediate between terminological niceties and the expectations of his audience. “Fear and Loathing’ is a more technical paper (still for a general audience), so compromises less.
What I see is a discipline maturing: linguisticians [sic] are throwing off outdated and never-applicable terminology. A text from 1985 might as well be talking about Phlogiston.
Even Chomsky 1956 ‘Three models’, which bases a lot of its case on analysing the be+en verb form, never uses “voice”; only “passives” or “passive transformation”.
This, actually, can be a useful development, if linguists (not the army type) adopt some other locution for passive [something] to the exclusion of voice. All cool people will then discuss linguistic niceties and “passive voice” would be a sure sign of crackpots.
Haspelmath has a consistent convention of using Capitalized Terms for language-specific terminology and lower-case terms for comparative concepts. Thus in English the Present Progressive is the usual way of expressing the present tense.
@Bathrobe: The earliest OED citation for the definition “a person who is skilled in the learning or use of foreign languages” is from 1582, and the earliest citation for “an expert in or student of language or (later) linguistics; a person who specializes in the structure or historical development of one or more languages; a philologist” is 1605, a quarter of a millennium before de Saussure was born.
Perhaps they mean “perimeter”. I have never encountered either in that sense.
Marja Erwin: I feel like many informal ways to emotionally voice one’s disapproval may reinforce prejudices against (or taboos about) people, their disabilities, animals, body parts, or body functions, so I try to say, for example, “_ is bad” as far as possible.
“an expert in or student of language or (later) linguistics; a person who specializes in the structure or historical development of one or more languages” from 1605? I would be interested to see the context, given that historical linguistics as we know it did not start until much later.
Here you go, I found the full source of the OED’s earliest citation in archive.org for you: “Languages” in Remains concerning Britain, by William Camden.
Incidentally OED helpfully defines “Realist” here as “A person who occupies himself or herself with things rather than words. Obs,” with the same Camden quotation as the first citation. (#2 is from 1613: “These men haue the smooth voice of Iacob & the rough hands of Esau. They are good linguists, but they are bad reallistes.” T. Powell, Serm.)
But people did study language (as opposed to knowing specific languages) half a millennium ago. The 1605 quotation’s context (which is too long to type in from Google Books) is a chapter on languages from a book by WIlliam Camden entitled Remains Concerning Britain. This chapter explains that the original language of Britain, so far as is known, is Welsh; that Latin was spoken in the island extensively, and left many words in Welsh, but was eventually displaced; that the language of the Scottish Highlands (or Hechlands, as he calls them in Scots) is of Irish origin; that English is an offshoot of the Saxon language of the Continent; that both [Low] Saxon and English-Saxon (as he calls it) are related to different varieties of German, as is shown by the closely related words for ‘and’; that a Germanic language is spoken in the Tauric Chersonese [Crimea] with many words in common with English; that although English has borrowed many words from Latin, its older variety was a language sufficient to express everything that needed to be said; and finally, though the author dare not affirm it himself, he reports that Joseph Scaliger believed that English was related to Persian, citing several common words. Surely this is linguistics.
He also tells us something useful while describing the ideas of his contemporary Thomas Smith, a spelling reformer, namely that in his day, though people still wrote dieth, lieth as well as dies, lies, the reading pronunciation of each was the same, monosyllabic and ending in /z/.
I just saw a genuine man-bites-dog story – a little internet squib that refers to the “passive voice” in order to make a political point and … correctly contrasts passive-voice constructions with the relevant active-voice counterpart. There may be other problems with the point being made (there are lots of other areas where statistics are sometimes kept on the basis of counting victims/incidents rather than counting victimizers/perpetrators where no one would think doing so was intended to obscure or diminish the culpability of the latter), but “passive voice” is used so correctly that no one could peeve about it. http://lilith.org/articles/when-you-use-the-passive-voice-for-gender-violence/
FWIW I am definitely of the opinion that there ought to be an English word in wide use for someone-who-does-linguistics-stuff other than “linguist,” but esp since “philologist” has become near-archaic (and was always a bit contested anyway) I’m afraid we don’t have a good candidate in actual use. Linguistician?
He also tells us something useful while describing the ideas of his contemporary Thomas Smith, a spelling reformer, namely that in his day, though people still wrote dieth, lieth as well as dies, lies, the reading pronunciation of each was the same, monosyllabic and ending in /z/.
Fascinating, I’m very glad to learn that!
‘Voice’ was, of course, a morphological category in Greek and Latin, but it had a loose semantic correlate, just as the morphological category of gender had a loose semantic correlate in sex. What objection can there be to using the term ‘passive voice’ in English / French / German etc. grammar to denote the compound verbal form that expresses much the same thing in those languages?
Fun fact: it isn’t called “voice” in German. I was just taught “the active” and “the passive”, without a term for what they both are; the first 9 ghits for passiv deutsch seem to do the same, the tenth is the Wikipedia article which calls them genera verbi and adds the impressive footnote “auch Verbalgenera, Verbgenera, Verbgeschlechter, verbale Genera, Genera, Diathesen, Handlungsformen, Aktionsformen, Zustandsformen oder Handlungsrichtungen genannt”.
In Russian the word is залог [zalóg] = za– + leg/log/lag/lezh/lozh ‘lie (down), recline’; the other meanings — ‘pledge,’ ‘deposit,’ ‘security,’ etc. — are easily derived from ‘something laid down (as a guarantee),’ but the development of the ‘voice’ meaning “is not clear” (to quote Herman’s Dictionary of Slavic Word Families).
What is the -tesis in Diathese, I have too little Greek to tell? (I assume the h in German is a spurious ornament, Danish doesn’t have it in its obscure but seemingly official term for voice).
θέση
διάθεση
Presumably the same thing as in English thesis (as well as synthesis and hypothesis), which is Ancient Greek θέσις (thésis), “a proposition, a statement, a thing laid down”, according to Wiktionary.
There is no spurious ornament, or, rather, to the extent that an ornament exists, it is not spurious; the th quite correctly represents the Greek letter θ (theta).
EDIT: Wiktionary does have diathesis, incidentally, and the second definition is “(grammar) voice (active or passive)”.
I would rather search for the source of залог by analogy with предлог (predlog, meaning prefix or pretext) or пролог (prolog, prologue), which obviously come from λόγος
My apologies to Greek. Since Danish doesn’t have dental fricatives, especially not in Anlaut, someone must have rationalized away the H. (But arguably it is an ornament in German since it makes no difference there either, and many of the TH’s in German are spurious, are they not?)
I would rather search for the source of залог by analogy with предлог (predlog, meaning prefix or pretext) or пролог (prolog, prologue), which obviously come from λόγος
The latter might, but does not appear relevant, while the former makes perfect sense as native Slavic* (i.e. “laid before”; cf. предлагать, предложить “to offer, to propose” and предложение “offer, proposal; sentence”).
[EDIT: I have no idea why the Russian term for “offer, proposal” also means “(grammatical) sentence”; I suspect they just happened to be formed in the same way with different intended meanings.
In modern Russian the “proposal” sense is commonly used as an abbreviation of “marriage proposal”, which is quite funny considering the grammatical sense – though I cannot recall any pun that involves it specifically.]
*) I’m not calling it “native Russian” because for all I know it may be of Church Slavonic origin
Vasmer says предлог is a calque of Greek πρόθεσις, but doesn’t say when it was created.
Melety Smotritsky in his 1619 Slavonic grammar uses zalog without any explanation as to the source of the word (he doesn’t explain the source of his terminology at all).
Interesting to know it goes back that far.
There were lots of spurious ones before the first common orthography was introduced in 1901. The few survivors almost all represent Greek θ; all are pronounced /t/.