Initial coordinators in technical, academic, and formal writing

« previous post | next post »

Yesterday, I quoted someone writing on the nanowrimo forum ("Also, check the back seat", 11/7/2009), who offered an apparently irrefutable argument in favor of "No Initial Coordinators" (NIC), the zombie rule that forbids us to begin a sentence with a conjunction such as and or but:

[Usage standards and grammar] are related but not identical. Grammar deals with categories such as parts of speech, and the logical rules of syntax for constructing sentences. Grammatically, conjunctions link words, phrases, or clauses. So from a grammatical standpoint, a sentence beginning with a conjunction is a fragment, and hence ungrammatical.

The lovely thing about this argument is the universality of its structure. We stipulate that the role of X is to perform functions A, B, or C; since D is not in our list, it follows logically that X cannot legitimately perform the function D.  To add to its lustre, this particular instance of the argument is self-refuting as well as circular, since some expressions of the "No Initial Coordinators" pseudo-rule include so in the taboo list.

Having conceded that initial coordinators are in common use, despite being (in his opinion) deductively ungrammatical, our grammatical zombie offers this generalization:

Although beginning a sentence with a conjunction is acceptable in fiction (there is wide agreement on this), it is not acceptable in technical, academic, or formal writing.

If generations of Hollywood movies have taught us anything, it's that you can't reason with zombies. That's why yesterday's post quoted Zombieland rule #22, "When in doubt, know your way out". But zombie logic from self-appointed authorities often lures innocent youth into the clutches of the undead, so I'm going to devote a few minutes this morning to examining the facts of initial-conjunction usage, in the hopes of inoculating a few I. Y.  against this particular strain of the zombie virus.

We're told that "beginning a sentence with a conjunction … is not acceptable in technical, academic, or formal writing".   The implication is that in well-written and well-edited writing of the specified kind, sentence-initial conjunctions should be absent or at least very rare,  rather like instances of ain't.

To find an example of "technical" writing, I turned to the most recent issue of one of the technical/scientific journals that I subscribe to, the Journal of the Acoustical Society of America, and looked at the first article in the table of contents, Derek C. Bertilone et al., "Ultra-wide sensor arcs for low frequency sonar detection with a baffled cylindrical array", JASA 126(5): 107-111, November 2009. The first instance of allegedly unacceptable behavior comes at the start of the second sentence (emphasis added)

Passive cylindrical sonar arrays are operated over the widest possible frequency range to exploit all available acoustic energy.  But it is a challenge to obtain acceptable performance at low frequencies due to high levels of ambient and platform-generated noises and poor bearing resolution

Clearly the American Institute of Physics, JASA's parent institution, remains free of this zombie virus.

OK, let's turn our attention to academic writing. Among the  publications that I subscribe to, perhaps the most clearly "academic" one is The Chronicle of Higher Education, which advertises itself as "the No. 1 source of news, information, and jobs for college and university faculty members and administrators". Since the main section of the Chronicle might be seen as too journalistic, I decided to turn to The Chronicle Review, a weekly insert that presents essays of a more formal and less journalistic character. In the most recent issue, the article that first caught my eye was Josh Fischman, "Global Warming Before Smokestacks" — where again, the first instance of "unacceptable" behavior comes at the start of the second sentence:

People have changed the planet's climate, warming the atmosphere by churning out greenhouse gases.

But that process didn't start during the Industrial Revolution. It began thousands of years ago, according to a controversial hypothesis, before anyone uttered the phrase "global warming."

Just to clarify that but is not the only coordinator that appears in sentence-initial position in such writing, a bit further along in the same essay we find

Within the past 10,000 years, humans began changing the planet, clearing land to grow food by cutting and burning forests. A map of archaeological sites shows signs of domesticated grains, and of cut-and-burned fields, that began 10,000 to 9,000 years ago in the Middle East. That practice spread north, east, and west until, by 5,700 years ago, it covered all of Europe and parts of Asia. And in Asia, by about 5,000 years ago, a good deal of land was cleared, flooded, and turned into rice paddies.

When I look around for an example of "formal" writing that's neither "technical" nor "academic", the first candidate in the array of books on my desk is Dwight MacDonald's collection of essays Against the American Grain: Essays on the Effects of Mass Culture.  Opening it, I find that it begins with an Introduction by the arch-prescriptivist John Simon. The second paragraph of this essay includes the sequence

It is his uncompromising yet utterly accessible, jargon-free, lavishly bequeathed individuality that made Dwight the universal critic he was.  And critic he was of everything he touched or was touched by.

Turning to MacDonald's own Preface, we need to read through to the third paragraph before finding an instance of an "unacceptable" sentence-initial coordinator:

Let it be admitted at once, as Dr. Edward Shils and other Panglosses of the sociological approach keep insisting, that mediocrity has always been the norm even in the greatest periods. This fact of life is obscured by another: when we look at the past, we see only the best works because they alone have survived. But the rise of masscult has introduced several new and confusing factors.

Could this "ungrammatical" licentiousness be a feature of earlier generations of intellectuals, a practice now purged from the formal writing of more sedate and serious modern thinkers? A bit further to the right on my desk, I find Richard Posner's recent A Failure of Capitalism: The Crisis of '08 and the Descent into Depression. The third paragraph of the first chapter begins

The flaw in this classical economic theory of the self-correcting business cycle is that not all prices are flexible; wages especially are not. This is not primarily because of union-negotiated or other employment contracts. Few private-sector employers are in the United States are unionized, and few non-unionized workers have a wage guaranteed by contract. But even when wages are flexible, employers generally prefer, when demand for their product drops, laying off workers to reducing wages.

In a later post, I'll take a look at the frequency of sentence-initial coordinators across genres and across time, and suggest what the results tell us about the causes and consequences of this zombie-virus strain.


  1. stripey_cat said,

    November 8, 2009 @ 11:15 am

    You'll find that most people with the old-fashioned "classical" education go wildly overboard on initial coordinators, as they're a common (almost every sentence) feature of Greek prose. After a few hours a day of Thucydides, or any similarly attention-absorbing author, you start to think in that syntax yourself. Of course, a lot of our prescriptivist rules come from analyses of Latin grammar, so there may be some tension between the Hellenists and the Romanists showing (both sides tend to inferiority complexes and the denigration of the other!).

  2. Ryan Denzer-King said,

    November 8, 2009 @ 11:36 am

    The problem I run into with some prescriptivists I know (though for the most part they aren't as zombie as those discussed here) is that they don't accept data. When I point out that plenty of respected writers break their "rules" of grammar, they simply say that either (a) (more rarely) that they don't respect those writers or (b) (more commonly) that the great ones always break all the rules, but that you have to learn the rules to break them, and the rules are still useful for students learning to write.

    How do you argue against someone who doesn't accept arguments based on what the majority of writers do, and doesn't accept arguments based on what a small number of great writers do?

    [(myl) People who refuse to participate in any form of rational discourse about grammar are what I'm calling "grammatical zombies", because you can shoot their arguments through the heart, but they just keep on coming. That's why you need to remember Rule #22: "When in doubt, know your way out". ]

  3. Troy said,

    November 8, 2009 @ 11:40 am

    Interesting point, stripey-cat. I only took a year of Ancient Greek, and it sure does have the effect you described. I find myself starting sentences with "Accordingly" now, something I never did before. Not only are initial coordinators common, they also frequently occur in groups. And so, therefore, we must accordingly agree that the demise of Greek education is to be more lamented perhaps than even the ongoing demise of Latin as far as instruction is grammar is concerned.

  4. CWV said,

    November 8, 2009 @ 12:57 pm

    I've met a lot of people who believe that initial coordinators are ungrammatical. Fortunately, most of them have been open to persuasion that NIC is not, in fact, a rule of grammar. (I find that citation to authority is vastly more effective on this point than citation to actual usage If a thick reference book says it's grammatical, that seems to be worth more than a thousand examples of Shakespeare, Milton, and Dickens actually using the language.) After accepting that NIC is not a rule of grammar, though, their fall-back position — and the one that I find in some ways more frustrating — goes something like this: "Even if it's true that NIC is not a rule of grammar, there are enough people out there who think that it is that we should follow it so as not to annoy or distract those people." In other words, when you live in a world of zombies, you follow zombie law.

    I usually try to convince them that the judicious use of initial coordinators can greatly improve the cohesion and persuasiveness of their writing, and that this benefit has to be weighed against the cost of angering a few zombie readers. But if they've lived their lives thinking that initial coordinators are ungrammatical, they've never discovered their benefits and are disinclined to give them much weight.

    Any suggestions on how to counter the "let's not piss off the zombies" argument (which, of course, applies not just to NIC, but to lots of other zombie rules as well)?

    [(myl) Arnold Zwicky, who has as many names for socio-grammatical phenomena as children do for types of Lego pieces, has called this the "Crazies Win" argument. He suggests that we decline to avoid offending grammatical crazies; but he notes that the ethics and etiquette are far from obvious, especially with respect to advising students. (In my opinion, this is where Rule #31, "Check the back seat", comes in.)]

  5. Spectre-7 said,

    November 8, 2009 @ 1:04 pm

    How do you argue against someone who doesn't accept arguments based on what the majority of writers do, and doesn't accept arguments based on what a small number of great writers do?

    Rule #22 above is certainly a good thing to keep in mind, but I consulted some of the more traditional resources on the subject, and I believe their advice could prove even more useful.

    "The Survival Command Center at the Pentagon has disclosed that a ghoul can be killed by a shot in the head, or a heavy blow to the skull."

    "Well, there's no problem. If you have a gun, shoot 'em in the head. That's a sure way to kill 'em. If you don't, get yourself a club or a torch. Beat 'em or burn 'em. They go up pretty easy. "

    I suspect that such methods are frowned upon in formal debate, but the line between caution and overkill blurs a bit when confronting the living dead.

  6. Janice Huth Byer said,

    November 8, 2009 @ 1:06 pm

    The nanowrimo writer's argument exemplifies a theory that the purpose of grammar rules is to protect not clarity but class status. By excusing fiction from rules for "technical, academic, and formal writing" the writer seems to classify all literature as informal writing. Ain't fair, that there.

  7. jolo said,

    November 8, 2009 @ 1:22 pm

    One of first classes I took to get a technical writing degree was entitled something like 'Style in Technical and Scientific Writing' and one of the first things I remember learning was that the NIC rule (not that we called it that) was bogus. My instructer's theory was that children were taught this at a young age to encourage them to write in longer, more complex sentences instead of short and choppy statements.

  8. theophylact said,

    November 8, 2009 @ 1:34 pm

    Part of the problem here is addiction to definitions. Who decided that "conjunction " was the right and only term for this structural element? "Coordinator" gives you a lot more leverage, of course, but who decided that the etymology of the word "conjunction" limited its application in the first place? A dictionary definition is not the QED, the snapper to end all argument, as so many posters to forums seem to think. As Lincoln is said to have remarked, "Calling a tail a leg doesn't make it one."

  9. Ran Ari-Gur said,

    November 8, 2009 @ 1:36 pm

    Language Log has previously settled prescriptivist questions, such as singular they, by examining Divine usage (see, and to a lesser extent NIC is a perfect use of that approach: of the 80 verses that make up the first three chapters of the Bible (WLC), all but 5 start with coordinators (74 v', meaning "and", and one ki, meaning "for"). Translations are harder to examine, because they don't always break up sentences along verse lines, but a skimming of the first three chapters of the KJV suggests it has a similar ratio: it removes a small number of NICs by using an adverb, and a larger number by separating verses by colons (which I assume is considered sentence-internal punctuation?) rather than periods, but conversely, it often splits a single verse into multiple sentences, and in that case the second one (almost?) invariably starts with "and".

    Really, it seems like it requires some poetic license not to start a sentence with a coordinator.

  10. Mr Fnortner said,

    November 8, 2009 @ 1:57 pm

    One, begging the question is the line of reasoning that says we should acknowledge that good writers are being correct when they write because they wouldn't have gotten to be good writers had they not been correct. And, two, I have always felt that definition by enumeration was the weakest form of definition, so that saying a conjunction is a word permitted to link A, B, C, and D is so open-ended that one could adjust the definition to include E and F, thereby adapting to any counter argument on the fly. That said, I agree with initial conjunctions. I simply believe linguists need a tighter definition than one that depends on a list, and a argument for correctness that doesn't depend on a survey of who's been using it.

    [(myl) The "list of permitted categories" approach was contributed by Tom L Waters on nanowrimo, the "zombie grammarian" who (I presume) is not a linguist. I mentioned the structure of the argument in this post simply to ridicule it. But thanks for agreeing that stipulated enumeration is a weak argument for excluding the things that are left out of the list.

    As for the role of surveys: If we want to determine the grammatical norms of some designated group of speakers or writers, how better to determine what these norms are than to examine the facts of what typical group members write or say?

    There are certainly other forms of argument that can be (and have been) tried. We can construct a model based on simple cases, and apply it to complex ones — but if its predictions are different from the actual facts of usage, this should make us doubt our model at least as much as it makes us question the correctness of common patterns of usage. We can do psychological or neurological experiments on subjects' reactions to different classes of examples — but again, if we interpret the experiments to mean that certain aspects of well-written and well-edited writing are almost always "ungrammatical", then we've got quite a bit of 'splainin' to do.

    Such arguments are certainly possible. I've tried to make one in certain cases of mis-negation. But I've never seen any attempt to determine the "correctness" of sentence-initial coordinators by such means, and I'm skeptical that any such argument could be constructed, other than as a joke.]

  11. ulyssesmsu said,

    November 8, 2009 @ 4:54 pm

    In writing, as in life, common sense must prevail. When you say "We're told . . . " not to begin sentences with initial coordinators, my first question always is, Told by who[m]? Who makes up these so-called "rules"? Is it the same "authorities" who tell us that we can't put prepositions at the end of sentences, or that we can't split infinitives? If so, those are authorities we need to ignore, because they obviously don't know what the hell they're talking about.

    I can't tell you how many times I've seen, in authoritative reference books, statements like "Everyone knows BY NOW that it's OK to put a preposition at the end of a sentence." But when one reads and listens to typical English prose today, it's quite obvious that NOT everyone knows this, because most people are still enslaved to these idiotic rules. If only any of us today could have the influence of Dryden (the originator of the no-preposition-at-the-end rule), whose nonsense has lasted now for almost 350 years!

    However, my primary point is that in technical, academic, and formal writing, there is NO SUCH RULE as "No NIC" or "No preposition at the end of a sentence" or "No splitting of infinitives." These are all "usage opinions" that have assumed the status of "rules" only by sheer force of repetition. But they defy common sense, and should have long since been rejected.

    How anyone who claims to know anything about writing can follow or, far worse, teach these foolish ideas is an enigmatic mystery beyond explanation.

  12. Non Naive Speaker said,

    November 8, 2009 @ 5:39 pm

    Divinely inspired translators use initial coordinators plenty. I just did a go-to-bed experiment and counted " And" in the Book of Mormon: more than 4400. And " But": more than 430.

  13. jean-pierre metereau said,

    November 8, 2009 @ 6:36 pm

    I'm so glad you gave me such a fine example of begging the question, an expression whose meaning seems to be forgotten by more and more people, if they ever knew it. And I know that this post leaves me open to the charge of peevology, but I can take it.

  14. Cheryl Thornett said,

    November 8, 2009 @ 6:37 pm

    But don't begin every other sentence with 'and', as the writer of one children's book did, quite literally. I loathed that book after the third reading, finally crossed out about half the 'ands', but still hated reading it. My son, alas, loved it. I had to excuse him as he was about 3 at the time.

    The correct thread this time, I hope.

  15. Scriptor Ignotior said,

    November 8, 2009 @ 6:40 pm




  16. Andrew (not the same one) said,

    November 8, 2009 @ 6:59 pm

    Interestingly, the NIC rule seems to conflict with another zombie rule. If one wishes to avoid initial 'but', an obvious move is to write 'however' instead, snce 'however is not normally classed as a conjunction (and this is frequently done); but there is a 'rule' upheld in some quarters that 'however' never comes at the beginning of a clause. (One can observe both rules, by refusing to use any adversative word as the first word of a sentence; but the way the two rules push in opposite directions is still striking.)

    [(amz) The conflict between NIC and Garner's Rule (advising against sentence-initial connective however) is one of the points of my 2006 posting "However,…"]

  17. Jerry Friedman said,

    November 8, 2009 @ 8:26 pm

    @CWV: Why shouldn't a modern usage book outweigh a thousand examples from Shakespeare and Milton? Shakespeare wrote "the most unkindest cut of all" and Milton wrote "Him who disobeys, Me disobeys." Examples such as those have no relation to modern grammatical English. Dickens, though, is more like it.

    [(myl) The value of 16th- and 17th-century examples is to show that this syntactic pattern has been consistently used for more than four centuries — in fact, more than 12 — and thus is not a culpable innovation to be shunned by those who wish to preserve the glories of etc. There are other patterns that began to be used a mere century or two ago, and some may choose to draw a line in the cultural sand to keep them back. But holding the line against sentence-initial coordinators is like holding the line against beer.]

  18. Mark F. said,

    November 8, 2009 @ 8:40 pm

    CWV has a good point. For a lot of people, the evidence of authority is the only evidence that counts here. I think the solution is for one of you to write a usage guide (perhaps it could be marketed as a grammar text) that made all of its pronouncements ex cathedra, with no justification beyond perhaps some undefended claims about clarity and strength. Giving evidence based on observed usage just brands you as a lax, anything-goes descriptivist. Don't say that NIC is a "rule" that Shakespeare (and everybody else) has ignored, just say that it is not a rule. Or ignore it entirely, but give examples of how you can use conjunctions at the beginnings of sentences in a discussion of how to structure an argument, or of avoiding sentence fragments. Perhaps you should say that the preferred position for an adverb with an infinitive having "to" is between the "to" and the verb, just to shift the position of the debate.

    Of course, if the goal is to help people understand that it's the usage of a language community that defines correctness, then this approach is counterproductive. But perhaps sometimes causes have to be recognized as lost. For those who can get it, there's Language Log. For those who insist on Authority, there's the Language Log Prescriptive Guide to Grammar.

  19. D.O. said,

    November 8, 2009 @ 10:59 pm

    In order to combat NIC zombies, I propose to look at the ultimate piece of formal writing in the U.S., the Constitution (other countries may choose for themselves). My source of the text is (somewhat insufficiently) wikisource. The original 1787 text is heavy on ands and buts after semicolons, some others follow colons with the first capital letter, but let's be strict. Only those ands and buts which follow a full stop count. Here we have:
    Article I, Section 7: But in all such Cases… Article II, Section 1: And they shall make a List of all the Persons… But in chusing the President… But if there should remain… Article IV, Section 1: And the Congress may by general Laws prescribe…

    There are no examples in the Bill of Rights. And Art. II, sec. 1 is in relevant parts overriden by 12th amendment (ratified 1804), which helpfully has: But in choosing the President… (only the spelling of choose has changed), And if the House of Representatives… But no person constitutionally ineligible…

    Moving forward:Amendment XIV (ratified 1868), Section 2: But when the right to vote at any election… Section 3: But Congress may by a vote of two-thirds… Section 4: But neither the United States nor any State… Amendment XXII (ratified 1951), Section 1: But this Article shall not apply to any person…
    Thus, if the encountered zombie is American, you have a good chance to escape by brandishing the Constitution and demanding that only an explicit interpretation to the contrary by the Supreme Court (which, I am sure, is also no shy in using sentence initial conjunctions) will change your point of view.

  20. uberVU - social comments said,

    November 8, 2009 @ 11:19 pm

    Social comments and analytics for this post…

    This post was mentioned on Twitter by languagelog: Initial coordinators in technical, academic, and formal writing: Yesterday, I quoted someone writing on the nanowrim…

  21. Stephen Nicholson said,

    November 9, 2009 @ 1:09 am

    Language Log: Continuing to both educate and entertain.

  22. Clare said,

    November 9, 2009 @ 1:33 am

    Don't you just ask them what they mean by 'rule'?

    They're going to have to concede pretty quickly that they're no longer describing the phenomena. So get rid of the "grammar deals with" talk (that's description) and agree to stick with normativity.

    OK, so some people think we ought not to use initial co-ordinators – fine. But ask why. What's the point of any utterance? Here are three contenders:

    (1) To communicate an idea clearly
    (2) To produce an aesthetically pleasing sentence
    (3) To show you know a convention

    Ask them for evidence that (1) can't be met. Of course in nearly every example you find it will be, but if (god forbid) they find an isolated example, concede the point and note it must be on on a case-by-case basis given all the crystal clear examples in evidence.

    Then concede them (2). If an IC is clunky, it shouldn't be used. But refer to T.S. Eliot, Goethe and whatever ancient Greeks come to mind. Concede that there's no disputing taste. Say it in Latin and make a friend.

    Then concede (3). But remind them that the world comprises leaders and followers. And note how weak this goal is compared with (1) and (2), especially (1).

    Then ask them for a summary of what you've agreed to and on the basis of this what should go into style guides. Then have an interesting discussion. :-)

  23. outeast said,

    November 9, 2009 @ 5:19 am

    Coincidentally, a snarkily marked-up version of a letter discussing editorial cutbacks at the Toronto Star is floating around just now (see here. I have every sympathy with copy-editors, being one myself, but this marked-up letter is essentially a parade of inconsistently applied S&W-style zombie rules and shibboleths.

    Most Strunk-and-Whitean of all is the way in which two passives are struck out with an injunction to 'use active verbs'… just a couple of sentences before the editor (unnecessarily) converts an active verb into the passive:)

  24. Tom Saylor said,

    November 9, 2009 @ 5:34 am

    "Although beginning a sentence with a conjunction is acceptable in fiction (there is wide agreement on this), it is not acceptable in technical, academic, or formal writing."

    Has anyone yet pointed out that this statement violates the rule it supports? In traditional grammar, "although" is classified as a conjunction–a subordinating conjunction, but a conjunction nonetheless. The distinction is immaterial to the schoolmarms: they'll rap your knuckles just as hard if you start a sentence with "Because" as they will if you start a sentence with "But".

  25. Graeme said,

    November 9, 2009 @ 5:53 am

    I'm a legal academic. Our prose is decidedly formal, technical and academic.
    And analytical, often in a mealy-mouthed way.

    Strike out all sentences beginning 'But' (or 'However') and you'd decimate most law review pieces. We are also led to believe it is better not to let sentences grow too long, hence ICs are inevitably common.

    My sense is people object to 'And', 'Or' and sometimes 'But' as ICs simply as they are short. Rarely do I hear objection to 'However' or 'Although' as ICs.

    Conversely, French legal writers, with less pretence to 'if, then, but' analytical styles write much longer and more flowing sentences, full of commas, semi-colons and digressions.

  26. Karen said,

    November 9, 2009 @ 6:57 am

    "Begging the question"is a horrible translation of "Petitio principi" and begs to be be misinterpreted as a plain English expression. If we can get away with "ad hominem" or "post hoc", why not just stick with "petitio principi"?

  27. Peter T said,

    November 9, 2009 @ 7:17 am

    I began writing intelligence analyses under a famously-strict grammarian. This was not one of his rules – and every piece had one or two sentences beginning with BUT or AND. More generally, how prescriptivist does one have to be to qualify as a zombie? Is, for instance, pointing out to the teenager that saying "like" three times in every sentence adds little to the conversation a symptom of early-onset undeadness?

    Consciousness of style comes early: three year old in bath assured me me today that "I won't squirt ya" [with water pistol]. I reply "You won't squirt ya, but you will squirt me". She says "Alright, I won't squirt YOU" (and squirts me).

    [(my) Linguistic zombiehood consists in the relentless expression of idiosyncratic "rules", presented not as personal stylistic preferences but as facts about what is linguistically "correct", indifferent to all appeals to logic, authority, and precedent.]

  28. Matthew Stuckwisch said,

    November 9, 2009 @ 8:12 am

    Could we maybe interpret a sentence initial and or but as an adverb? It seems to me that would clear up a lot of the hang-ups that some people have as well as making it fit into even the oddest out-there views on English grammar.

  29. Ken Brown said,

    November 9, 2009 @ 9:35 am

    Your zombie grammarian is using the word "grammar" differently. What he means by "usage" is the way the language actually works, what you call "grammar" (or maybe "syntax").

    What he means by "grammar" is a set of arbitrary rules about language which schoolkids are taught by rote and are expected to be able to reproduce as a sort of proof that they spend all those years in school. Its not really the study of language at all, more an entry ritual into a certain social class. The point of teaching it is so that some will pass and others fail.

    I can sympathise with him – I was taught mathematics that way. No explanation or theory, just repetition of mechanical methods to answer trite questions. Not a good way to learn maths at all. But they seemed to think it important for some reason.

  30. dwmacg said,

    November 9, 2009 @ 10:09 am

    As an aside, am I the only one who initially read Rule 22 as an example of the way-construction (along the lines of "fight your way out")?

  31. Roger Lustig said,

    November 9, 2009 @ 10:13 am

    Along with grammar, syntax and usage there's style. And has suddenly sprung up to show us just how much fun it can be to prescribe this stuff.

    If it hurts when you laugh, don't follow the link.

  32. Jerry Friedman said,

    November 9, 2009 @ 11:40 am

    @myl: I've seen people attack constructions they don't like as newfangled, in which case showing that they go back centuries makes sense, but I don't recall seeing that attack on sentence-initial coordinators. After all, as you've now logged here in detail, they're plentiful in translations of the Bible. And CWV, who I was disagreeing with, didn't say he or she was responding to such an argument. I got the impression CWV was saying, "If it's good enough for the greatest users of English, it's good enough for us," and that's what I was rebutting.

    The "zombie rule" against SICs that you quoted was based on a belief that they're objectively wrong. Citing anyone's usage of SICs, even the prescriptivist's own, doesn't respond to the argument, any more than my sign errors in physics lectures justify my students' sign errors on their homework. Pointing out what's wrong with the rule and its supposedly objective basis, as you did in your original post, strikes me as a much better response.

  33. Jerry Friedman said,

    November 9, 2009 @ 11:42 am

    @Roger Lustig: Hi, it must be what, almost thirty years?

  34. CWV said,

    November 9, 2009 @ 12:57 pm

    My point was simply (1) that many people find it to be a persuasive refutation of the argument "X is ungrammatical" to show that the best writers in English have used X in polished, edited prose in an unbroken chain going back for centuries, but (2) the sorts of people who believe in NIC are less likely to be persuaded by this sort of evidence than they are to be persuaded by a passage in a reference book written by a self-proclaimed authority on English (even if the basis for this authority's opinion is the same evidence mentioned in (1) above).

  35. Jerry Friedman said,

    November 9, 2009 @ 3:30 pm

    Okay, then what you've found isn't surprising, because I think most people who have an opinion on SICs would already know that they appear frequently in the King James translation. So examples in Shakespeare and Milton wouldn't convince them of anything. Its the continuation of that chain of polished prose to the present day that might convince them, and the reference books you cite probably go by that continuation.

    On the other hand, I wouldn't be surprised to see SICs condemned as an archaism (but I don't remember ever seeing that).

  36. Mr Fnortner said,

    November 9, 2009 @ 4:34 pm

    @myl, thank you for the considered reply. Had you a unit similar to morpheme that was enlargeable in a well-behaved way, and that was rigorously defined, so that it began life as a word, enlarged gracefully to phrase, then to clause and on to sentence, you could define a conjunction as a word, or part of speech that connected two or more like units. With such a named unit, you would no longer need to rely on petitio principi (thank you, Karen), enumeration, or surveys.

  37. Greg Morrow said,

    November 11, 2009 @ 12:14 pm

    Guy Deutscher's "The Unfolding of Language" (which I found fascinating) suggests tentatively that there's some evidence that the only technological innovation that modern languages have compared to the most ancient writings we have available is a more sophisticated coordination apparatus.

    I.e., ancient narratives are "I did this. And then I did that. And then I did something else. And then I went somewhere new and did something new." The structure is repetitive (for ease in memorization, perhaps).

    Modern languages seem to use a richer vocabulary with less repetition to narrate the same sequence of events.

    If languages have been driven to develop more sophisticated coordination, e.g. by having a lot more written material around, and a lot less necessity to make passages easy to memorize, perhaps NIC is a result of the same impulse.

    [(myl) Confining ourselves for the moment to western traditions, this idea only works if you consider e.g. Plato to be "modern". Parataxis has been going in and out of style for at least a millennium and a half.]

  38. Greg Morrow said,

    November 11, 2009 @ 12:17 pm

    Also, if the notion is true, then there's not much point going back to the Hebrew Bible and the like for anti-NIC ammunition; NIC would have been developed precisely because the ancient narrative structure would have been considered a flaw to be avoided.

  39. Dan Cooper said,

    March 9, 2010 @ 1:08 pm

    Recently having been criticized for my use of ICs by an apparently crusty zombie, I was reminded of another sort of zombie rule about the construction, "one of those who," followed by a noun, the (number) agreement of which is a grammatical trouble spot. I was told that particular subject is addressed here somewhere, but it is remarkably well hidden, if present.

    The zombie rule on this that is frequently offered, is that the subsequent verb must agree with the subject, "one" despite the fact that virtually every source one consults, says otherwise. This is supposed to be a controversy, but seemingly none of the acknowledged authorities fall in with the so-called notional view, that agreement should be singular, with the singular noun, "one." Where are all the "authorities" who favor this construction? If there are none, then there really isn't much of a controversy, is there? And the zombies who profess the notional viewpoint as the law, can cite what sources again?

  40. Dan Cooper said,

    March 9, 2010 @ 1:14 pm

    Sorry. Mea culpa. I should have said "one of those who," followed by a VERB, the (number) agreement of which is a grammatical trouble spot.

RSS feed for comments on this post