Weak Interruptions - Trask et. al.
|
|
Thread rating:  |
rolleston - 01 Jan 2004 15:57 GMT L. Trask and others make use of the words `weak interruption' to define when bracketing commas may be used. Is this just a syntactic definition?
[Trask]
The rule is this: a pair of bracketing commas is used to mark off a weak interruption of the sentence, that is, an interruption which does not disturb the smooth flow of the sentence.
Smooth flow? I'm not sure what that means in precise terms. The author later states:
In each case a weak interruption has been set off by a pair of bracketing commas...in every one of these examples, the weak interruption set off by bracketing commas could, in principle, be removed from the sentence, and the result would still be a complete sentence that made good sense.
Syntax: complete sentence. Semantics: good sense. Now, this is not necessarily a definition; it may be a comment that does not apply to every instance of bracketing comma use. But then he remarks:
This is always the case...and you find you can't remove those words without destroying the sentence, you have done something wrong.
Now the problem. He later writes:
There are a number of common words which typically introduce weak interruptions containing complete sentences. Among the commonest of these are although, though, even though, because, since , after, before, if, when and whenever.
Now, please consider this:
If I win the lottery, I shall be rich.
I feel that the comma is used in an acceptable way. But `If I win the lottery' cannot, surely, be viewed as a weak interruption. If we remove it, the sentence may make sense, but the wrong sense. This will be the case for most instances in which if-conditions are expressed. The interruption is in no sense weak.
Comments appreciated,
R.
John O'Flaherty - 01 Jan 2004 16:48 GMT >L. Trask and others make use of the words `weak interruption' >to define when bracketing commas may be used. Is this just [quoted text clipped - 40 lines] >This will be the case for most instances in which if-conditions are >expressed. The interruption is in no sense weak. Well, he said, in effect, that 'if' and like words typically introduce weak interruptions. He didn't say its presence implies one. I think he meant that, given that you have such an interruption, it wouldn't be unusual to introduce it with an 'if', if I've understood it right.
-- john
rolleston - 01 Jan 2004 23:30 GMT > Well, he said, in effect, that 'if' and like words typically introduce > weak interruptions. He didn't say its presence implies one. I think he > meant that, given that you have such an interruption, it wouldn't be > unusual to introduce it with an 'if', if I've understood it right. My argument is that it is unusual for an if-condition to be a weak interruption, but usual, or at least acceptable, for commas to delimit it. If the interruption is not weak, you have to find some class of comma usage to which this instance can be assigned. None of the other classes in Trask's book will accept it.
The principal reasons for using commas with if-conditions, as I view things, are:
(i) To highlight the logical structure of the sentence.
(ii) To avoid problems of ambiguity. If the if-condition goes first, and the apodosis is not introduced by `then', then there may be cases where words at the end of the first get confused with words at the beginning of the second.
E.g., If I say something strange things will happen.
R.
CyberCypher - 01 Jan 2004 17:03 GMT rolleston@onetel.net.uk (rolleston) wrote on 01 Jan 2004:
[...]
> Now the problem. He [Trask] later writes: > [quoted text clipped - 12 lines] > This will be the case for most instances in which if-conditions are > expressed. The interruption is in no sense weak. Removing the conditional clause leaves a "good" sentence and "a complete sentence that [makes] good sense", as Trask says [1]. "I shall be rich" is only a prediction, and the conditional, it seems to me, is a weak interruption simply because it is not explanatory. The speaker is not rich, and winning the lottery is a virtually unreal condition.
What about a "when" clause, eg, "When my inheritance is transferred to my bank account tomorrow, I shall be rich"? This is not a prediction but a statement of fact. He does say that "if" and "when" "typically introduce weak interruptions containing complete sentences". That should mean that they don't always do so.
He also gives an example that seems very iffy to me:
"Bracketing commas always come in pairs, unless one of them would come at the beginning or the end of the sentence, and they always set off a weak interruption which could in principle be removed from the sentence:
My father, who hated cricket, always refused to watch me play."
It seems to me that this example is not supportive of his case. Without the bracketed phrase, the sentence does not make "good" sense. I suppose it could, in principle, be removed if the context made it clear that what his father refused to watch him play was cricket and that why he refused to watch was that he hated cricket, but that relies on the entire discourse and not the sentence alone. Here is an example of a remaining sentence that is grammatical and makes the wrong sense:
My father always refused to watch me play.
I think you've got a good point here. It appears that Mr Trask has not given enough thought to the importance of semantics and the effect the removal of what he labels as "always . . . a weak interruption" might have on the entire discourse.
[1] http://www.cogs.susx.ac.uk/local/doc/punctuation/node13.html
 Signature Franke: EFL teacher & medical editor.
rolleston - 01 Jan 2004 23:47 GMT > > If I win the lottery, I shall be rich. > [quoted text clipped - 4 lines] > speaker is not rich, and winning the lottery is a virtually unreal > condition. I have to disagree here. A condition is an essential part of most conditionals. The truth of the main clause is critically dependent on the condition. `I shall be rich' on its own may be a prediction, but it is a bold unqualified one. I concede, though, that the main clause of some conditionals may be better able to stand on its own, e.g., where the word `would' occurs. In some cases the presence of an unexpressed condition may be inferred from the context.
Incidentally, I think that conditionals break the claim that subordinate clauses can be distinguished from main clauses because they depend on them for their meaning.
R.
CyberCypher - 02 Jan 2004 01:38 GMT rolleston@onetel.net.uk (rolleston) wrote on 02 Jan 2004:
>> > If I win the lottery, I shall be rich. >> [quoted text clipped - 7 lines] > I have to disagree here. A condition is an essential part of > most conditionals. Shouldn't this "most" be "all" in order for this claim to be consistent with your next sentence and your final sentence?
> The truth of the main clause is critically > dependent on the condition. 'I shall be rich' on its own may [quoted text clipped - 7 lines] > subordinate clauses can be distinguished from main clauses because > they depend on them for their meaning.
 Signature Franke: EFL teacher & medical editor.
rolleston - 02 Jan 2004 15:32 GMT > > I have to disagree here. A condition is an essential part of > > most conditionals. [quoted text clipped - 8 lines] > > subordinate clauses can be distinguished from main clauses because > > they depend on them for their meaning. No. First I address the general situation, in case your rebuttal is aimed at that. Then I discuss my particular example drawn from the set of `most conditionals'. The final point does not depend on `most' or `all'. The point is that the rule for identifying dependent clauses must apply in all cases. I merely need to show that it does not do so in some cases, i.e., for most conditionals.
R.
CyberCypher - 02 Jan 2004 16:37 GMT rolleston@onetel.net.uk (rolleston) wrote on 02 Jan 2004:
>> > I have to disagree here. A condition is an essential part of >> > most conditionals. [quoted text clipped - 11 lines] > No. First I address the general situation, in case your rebuttal > is aimed at that. I didn't think I was involved in a debate. I was just asking a question.
> Then I discuss my particular example drawn from > the set of `most conditionals'. The final point does not depend on > `most' or `all'. The point is that the rule for identifying > dependent clauses must apply in all cases. I merely need to show > that it does not do so in some cases, i.e., for most conditionals. Okay, I can buy the idea that the "condition is an essential part of most conditionals". That's a generalization that applies in "most" but not "all" cases.
Can you give me an example of a conditional in which the condition is not an essential part? I'm assuming that there are two types of conditionals here, explicitly stated conditionals of the type "If X, then Y", and implicit conditionals in which the "If X" portion is understood, has been previously stated, or is otherwise known to the speaker/writer and the listener/reader. Can there be a conditional without an explicit or implicit condition?
I suppose that you're invoking some rule or other that I am not aware of when you say that a syntactic relationship (clausal subordination) is voided by semantic necessity (In the sentence "If X, then Y", Y depends on X for meaning).
I don't want to argue about which takes precedence, syntax or semantics, but I don't know enough about the theory of how these two aspects of language interact such that the semantic necessity of the syntactically dependent, ie "subordinate", clause somehow makes it impossible to identify the syntactically dependent clause. The "if" clause requires the the "then" clause because it cannot stand alone as a "complete sentence" (given the appropriate definition of a complete sentence) [=syntactic dependency], but the "then" clause needs the "if" clause because it cannot stand alone as a "complete idea/thought" (given the appropriate definition of a complete idea/thought) [=syntactic dependency].
In other words, both clauses are dependent and neither one is independent. But this seems to be mixing up levels. I agree with what you say about each clause needing the other, but it doesn't make sense to me to argue that semantics can affect syntax in that way.
Is the following the kind of main clause containing "would" that you meant when you said "I concede, though, that the main clause of some conditionals may be better able to stand on its own, e.g., where the word 'would' occurs. In some cases the presence of an unexpressed condition may be inferred from the context"?
I would have the rain fall between 3:00 and 5:00 a.m.
 Signature Franke: EFL teacher & medical editor.
rolleston - 03 Jan 2004 02:15 GMT > I didn't think I was involved in a debate. I was just asking a > question. Sorry, probably worded too strongly.
> Okay, I can buy the idea that the "condition is an essential part of > most conditionals". That's a generalization that applies in "most" > but not "all" cases. I put the word `most' in as a weakener just in case some bright spark happened to come up with a counterexample. But just to clarify things: I'm not claiming that a conditional can be conditional if there is no implicit or explicit condition. But there are some conditionals from which the condition can be removed without changing the meaning too much, at least in an informal sense. For example, if the condition is always true. I think there may also be some sentences that are conditional in form but are not really conditionals. You might argue that the case where the condition is always true is of this type. But there may be other examples. So, when I say `most' conditionals, I'm just trying to be cautious. It's very difficult to make absolute statements about English because there are always counterexamples lurking.
> I suppose that you're invoking some rule or other that I am not aware > of when you say that a syntactic relationship (clausal subordination) > is voided by semantic necessity (In the sentence "If X, then Y", Y > depends on X for meaning). I don't think so. It's just that I've seen the semantic dependency rule given as a way of identifying the correct syntactic part. It's not a very good rule as it is of doubtful accuracy. If you believe the rule too strongly your syntax may end up ruined. R.
Odysseus - 01 Jan 2004 18:38 GMT > Now, please consider this: > [quoted text clipped - 5 lines] > This will be the case for most instances in which if-conditions are > expressed. The interruption is in no sense weak. I don't see this as an example of "interruption" because there's only one comma, not a "bracketing" pair as in "By this time next year, if I win the lottery, I shall be rich."
 Signature Odysseus
rolleston - 01 Jan 2004 23:15 GMT > I don't see this as an example of "interruption" because there's only > one comma, not a "bracketing" pair as in "By this time next year, if > I win the lottery, I shall be rich." When the interruption occurs at the beginning or end of a sentence, one of the commas may be dropped. The sentence above illustrates the point.
R.
CyberCypher - 02 Jan 2004 01:30 GMT rolleston@onetel.net.uk (rolleston) wrote on 02 Jan 2004:
>> I don't see this as an example of "interruption" because there's >> only one comma, not a "bracketing" pair as in "By this time next [quoted text clipped - 3 lines] > end of a sentence, one of the commas may be dropped. > The sentence above illustrates the point. I think that "I shall, if I win the lottery, be rich" would be an illustration of the point that Odysseus contests here. Lopping of the intial adverbial phrase alters the sentence structure so much that it fails to demonstrate anything of the sort. By placing the conditional at the beginning of the original sentence, one moves it from the middle of my example sentence without eliminating a qualifying phrase ("By this time next year", which is not there) and without changing the meaning of the sentence at all.
Trask did say that "'if' typically introduces a weak interruption". Now that he's clarified in another post that "weak interruption" has no concrete definition beyond, presumably, an unnecessary phrase or clause that is bracketed by commas and does not significantly interrupt the flow of the sentence, and that the term was coined ad hoc to avoid adding a 5th type of comma so as not to unduly confuse weak punctuators, the idea seems insufficiently worked out and, in the end, confusing even to skillful punctuators. I can see the difference between what he calls an abrupt interruption and a weak interruption, but I think neither of these terms is sufficient for his analysis. The abrupt interruption is more like an ejaculation and a digression from the main idea in the sentence while the weak interruption is complementary. There may already be adequate terminology to describe the differences between them.
 Signature Franke: EFL teacher & medical editor.
rolleston - 02 Jan 2004 15:53 GMT > I think that "I shall, if I win the lottery, be rich" would be an > illustration of the point that Odysseus contests here. Lopping of the > intial adverbial phrase alters the sentence structure so much that it > fails to demonstrate anything of the sort. I'm not sure what you mean here by initial adverbial phrase. `I shall'? I would say that it is not quite what you claim. This sentence provides an example of a real interruption, in that `if I win the lottery' breaks `I shall be rich' into two parts. Neither `I shall' nor `be rich' is complete in itself. Neither has been joined on with a single bracketing comma as the glue.
Your sentence does raise an interesting issue. Are the following identical in meaning:
(i) I shall, if I win the lottery, be rich. (ii) I shall be rich, if I win the lottery. (iii) If I win the lottery, I shall be rich.
The comma is not necessary for (ii). My intuition is that (ii) and (iii) are similar meaning. The condition in (i) strikes me as more of an afterthought, though there may be no formal difference.
R.
CyberCypher - 02 Jan 2004 16:42 GMT rolleston@onetel.net.uk (rolleston) wrote on 02 Jan 2004:
>> I think that "I shall, if I win the lottery, be rich" would be an >> illustration of the point that Odysseus contests here. Lopping of [quoted text clipped - 3 lines] > I'm not sure what you mean here by initial adverbial phrase. > `I shall'? No, I foolishly snipt that sentence:
"By this time next year, if I win the lottery, I shall be rich."
The initial adverbial phrase is "By this time next year".
> I would say that it is not quite what you claim. > This sentence provides an example of a real interruption, in [quoted text clipped - 14 lines] > in (i) strikes me as more of an afterthought, though > there may be no formal difference. In writing, they all seem identical in meaning to me, but, in speech, there might be a nuance to each structure, depending upon tone, stress, pitch, etc.
I don't see, though, how something said in midstream could be construed as an afterthought. That seems appropriate for # (ii).
 Signature Franke: EFL teacher & medical editor.
rolleston - 03 Jan 2004 02:24 GMT > > (i) I shall, if I win the lottery, be rich. > > (ii) I shall be rich, if I win the lottery. [quoted text clipped - 7 lines] > I don't see, though, how something said in midstream could be > construed as an afterthought. That seems appropriate for # (ii). Midstream but not midthought. It's all based on impression and feel. If I were constructing a sentence I would try to avoid (i). It's messy. So it strikes me as the sort of sentence that has been adjusted after it was initially conceived, but whilst being spoken.
R.
CyberCypher - 03 Jan 2004 02:34 GMT rolleston@onetel.net.uk (rolleston) wrote on 03 Jan 2004:
>> > (i) I shall, if I win the lottery, be rich. >> > (ii) I shall be rich, if I win the lottery. [quoted text clipped - 13 lines] > sentence that has been adjusted after it was initially > conceived, but whilst being spoken. I wrote that sentence to illustrate for some other poster how an if- condition clause could be bracketed by 2 commas without the initial time adverbial phrase "By this time tomorrow". I would avoid (i) too. It's quite inelegant.
 Signature Franke: EFL teacher & medical editor.
Mike Lyle - 01 Jan 2004 18:59 GMT > L. Trask and others make use of the words `weak interruption' > to define when bracketing commas may be used. Is this just > a syntactic definition? [...]
> Now, please consider this: > [quoted text clipped - 5 lines] > This will be the case for most instances in which if-conditions are > expressed. The interruption is in no sense weak. And it's in no sense an interruption, either: if such conditionals didn't come in pairs they wouldn't mean anything. If I haven't misunderstood you, you seem to have made a jump from "bracketing commas" to "all commas".
(I don't really see why the expression "bracketing commas" needs to be used, since -- as we have seen in a recent thread -- a bracket is one thing, and a comma another, but either may be used to mark a parenthesis. There seems to be interference from American layman's language, in which "parenthesis" is unhelpfully used to mean what over here is called a "bracket", in much the same way as "period" has there come to mean not "sentence" but "full stop".)
Mike.
Simon R. Hughes - 01 Jan 2004 22:45 GMT >> L. Trask and others make use of the words `weak interruption' >> to define when bracketing commas may be used. Is this just [quoted text clipped - 23 lines] > here is called a "bracket", in much the same way as "period" has there > come to mean not "sentence" but "full stop".) The first half is a conditional clause, made so by the "if". Standard comma rules dictate that a conditional clause coming first in the period is separated from the main clause by a comma.
It's a bad example of a parenthetical phrase appearing first in the sentence.
 Signature Simon R. Hughes
rolleston - 02 Jan 2004 15:39 GMT > > I feel that the comma is used in an acceptable way. But `If I win > > the lottery' cannot, surely, be viewed as a weak interruption. If [quoted text clipped - 6 lines] > misunderstood you, you seem to have made a jump from "bracketing > commas" to "all commas". I could have made the stronger claim, and I should have done. I agree with the first part of what you say. But I am not making the jump - it is in the book. This kind of use is dealt with under `bracketing commas', and nowhere else, as far as I can tell.
R.
Larry Trask - 01 Jan 2004 19:27 GMT > L. Trask and others make use of the words `weak interruption' > to define when bracketing commas may be used. Is this just > a syntactic definition? The term has no precise definition. It was merely the best label I could think of when I was writing my book. My idea was to find some way of distinguishing the use of bracketing commas from the use of dashes.
Consider one of my examples:
"The destruction of Guernica -- and there is no doubt that the destruction was deliberate -- horrified the world."
Here the interruption is so abrupt that bracketing commas are out of the question.
> [Trask] > [quoted text clipped - 3 lines] > > Smooth flow? I'm not sure what that means in precise terms. There is no possibility of giving a mathematical-style definition. But the Guernica example shows that there exists a real difference between abrupt interruptions and what I am calling "weak interruptions".
> The author later states: > [quoted text clipped - 29 lines] > This will be the case for most instances in which if-conditions are > expressed. The interruption is in no sense weak. You have a point, but I made this decision quite deliberately.
Please bear in mind that my book (The Penguin Guide to Punctuation) is written for readers who are very uncertain punctuators. It is meant to be useful to such readers. It is not aimed at skilful punctuators who are looking for guidance on subtle points, and it is not a scholarly treatise.
Accordingly, I made every effort to keep the book easy to follow. Another guide to punctuation currently in print distinguishes fourteen uses of the comma. My view is that a reader encountering a blizzard like this will find his heart sinking, and will very likely give up.
I therefore determined to keep the comma chapter as simple as possible. Originally I recognized only three uses of the comma, but a colleague who read a draft persuaded me to add a fourth, the one involving deletion of repeated material. I'm still not sure this addition was a wise decision.
I deliberately decided to include adverbial subordinate clauses under the rubric of weak interruptions. As you point out, this classification is questionable, but I decided that it was preferable to recognizing yet another distinct use of the comma.
If you are a skilful punctuator, this book will be of little use to you. But I was writing for the hundreds of millions who have no idea how to use commas.
And I might add that the comma chapter was the hardest chapter to write.
Larry Trask larryt@sussex.ac.uk
rolleston - 01 Jan 2004 22:51 GMT <interesting remarks snipped>
Thankyou for your helpful response. My point was probably a rather minor one, but what you have said has made things much clearer. I never know in these situations if I've made some horrible error. I remember an occasion when I had managed to convince myself and a friend that I'd found a major error in Papadimitriou's Computational Complexity. We were just about to contact him when I realised I'd made a wholly trivial mistake. Even now, I cannot read the proof in question and work out what on earth I was thinking. And if I can make an error in that area, where I am more knowledgeable, there is no doubt that I can err with regard to punctuation. So I'm pleased not to have got things completely wrong on this one occasion.
> And I might add that the comma chapter was the hardest chapter to > write. A few claims have been flung about recently. Some people seem to assert, for example, that punctuation usage is of infinitesimal difficulty, that schoolchildren can be taught it in a day, that anyone misusing commas must be stupid, and so on. I really don't believe such claims.
Once again, thanks,
R.
Mike Lyle - 02 Jan 2004 12:39 GMT > > L. Trask and others make use of the words `weak interruption' > > to define when bracketing commas may be used. Is this just [quoted text clipped - 4 lines] > way of distinguishing the use of bracketing commas from the use of > dashes. I think it's a very useful term, and that you've made a good case for it.
My only objection (stated upthread) is to the use of the extression "bracketing commas". I know "bracketing" is regularly used in a figurative sense, such as when a ship undergoes misses from shells on either side; but in the punctuation context I'm not quite comfortable with it, since there "bracket" has a literal meaning. For example, I might be tempted to distinguish "*full-stop question-marks" and "*comma question-marks", and I imagine many people would know what I meant; but I'll bet some would find it perverse.
Then again, "parenthetic" isn't the most layman-friendly of words, so perhaps you were in a cleft stick.
[...]
> Accordingly, I made every effort to keep the book easy to follow. > Another guide to punctuation currently in print distinguishes fourteen > uses of the comma. My view is that a reader encountering a blizzard > like this will find his heart sinking, and will very likely give up. [...]
It would be daft to disagree. I haven't seen the book, but I'm sure not only...er... greengrocer's will profit from it.
Mike.
Larry Trask - 02 Jan 2004 19:21 GMT > My only objection (stated upthread) is to the use of the extression > "bracketing commas". I know "bracketing" is regularly used in a > figurative sense, such as when a ship undergoes misses from shells on > either side; but in the punctuation context I'm not quite comfortable > with it, since there "bracket" has a literal meaning. The label 'bracketing commas' is well established here in Britain. But bear in mind that parentheses (those round things, like these) are commonly known as 'brackets' in Britain. To be finicky, they are 'round brackets', while the other things are 'square brackets' and 'angle brackets', but unmodified 'brackets' is usual.
[LT]
> > Accordingly, I made every effort to keep the book easy to follow. > > Another guide to punctuation currently in print distinguishes fourteen [quoted text clipped - 4 lines] > It would be daft to disagree. I haven't seen the book, but I'm sure > not only...er... greengrocer's will profit from it. The Penguin Guide to Punctuation is not distributed in the US. The American branch of Penguin declined to take the book, apparently because they considered it "too British", and they chose instead to re-issue the venerable Mentor guide. This is silly, since I'm American, and I wouldn't have needed a day to revise the book for an American readership.
But the book is on sale in Canada, and you can order it from the British and Canadian arms of Amazon.
Larry Trask larryt@sussex.ac.uk
rolleston - 03 Jan 2004 01:24 GMT > But the book is on sale in Canada, and you can order it from the > British and Canadian arms of Amazon. I wonder if you could recommend a fairly comprehensive modern English grammar book. For some reason I've ended up knowing more about Greek, French and Latin grammar than the grammar of my own language. This relative adverb thing has revealed some hideous gaps in my basic knowledge of the language.
Thanks,
R.
rolleston - 01 Jan 2004 21:09 GMT > Now, please consider this: Uh! I wish I hadn't written that. Looking at it again, it looks so pompous. And then there's that inelegantly repeated `now'.
> I feel that the comma is used in an acceptable way. But `If I win > the lottery' cannot, surely, be viewed as a weak interruption. If > we remove it, the sentence may make sense, but the wrong sense. ^ And that's an incorrectly used gapping comma.
Nevertheless, and despite the stylistic flaws, I think the argument is still worth thinking about. I thought I might just add a little to it. Some authors use the `makes good sense' rule to allow commas for non-defining relative clauses, and to forbid them for the other case. That is a case where the syntax rule alone is not enough. But, as I have said, neither rule works for if-conditions. And I suspect that is the case for some of the other words in Trask's list.
R.
ps To say `syntax rule' and `good sense rule' is probably to extend things a little beyond what has been claimed.
|
|
|