grammar, syntax and, semantics
Neil Ramsay
I came across a proofreading checklist that contained, among other
things, the following items:
- grammar errors
- syntax errors
- semantic errors
If you are asked to look for syntax errors, what exactly should you be
looking for? And how exactly would this differ from what you would
look for if you’re asked to check for grammar errors?
I would be very grateful if anyone can shed some light on this for me.
Best regards,
Neil Ramsay
Stephen A. Carter
Last week, a client suggested a change to a translation of mine.
I had written this:
"This feature automatically switches off the power when the machine has
been left unattended for a specified interval."
The client wanted to change it to this:
"This feature automatically switches off the power for a specified
interval when the machine has been left unattended."
That would have been a syntax error, as I understand it. (The original
Japanese was something along the lines of 「一定時間放置後のオートオフ」.)
--
Stephen A. Carter
sca...@hticn.com
Nagoya, Japan
Jens Wilkinson
> That would have been a syntax error, as I understand it. (The original
> Japanese was something along the lines of 「一定時間放置後のオートオフ」.)
>
Yes, I agree. That definitely have been would a syntax error.
And I think a semantic error would be something like:
"An elephant is a large African reptile."
What's a little bit unclear to me is the difference between a
grammatical error and a syntax error. Maybe grammar is a bit broader
than syntax, and would include things like number and tense. So for
example:
"I ate one apples."
I thinks that would be a grammatical error.
--
Jens Wilkinson
Neo Patwa (patwa.pbwiki.com)
Mark Spahn
Excuse me, but I think there's something defective
with the original assignment. As I understand it,
language has three features: semantics (= meaning),
syntax (= grammar, which words fit with which), and
pragmatics (= social relationships, e.g., whether
one does or does not use "sir" when addressing
someone; "pragmatics" sure is a misleading term
of art). That is, "syntax" is just a highfalutin'
word for "grammar". If anyone knows of a nontrivial
distinction between syntax and grammar, do tell.
In the example Stephen has given above, the
error is one of semantics: the meaning is different
depending on whether "for a specified period"
modifies "left unattended" (intended meaning)
or modifies "switches off the power" (unintended
meaning). Both sentences are grammatically correct,
but they do not express the same meaning.
Incidentally, a manual I have at hand for a
calculator with an auto-shutoff feature uses
the wording "[This feature] turns off the [calculator]
automatically after about five minutes without any
activity." Hmmm, 放置 = without any activity.
-- Mark Spahn (West Seneca, NY)
Stephen A. Carter
> Hmmm, 放置 = without any activity.
In the case of my machine, that translation would have been a semantic
error. Unlike my machine, though, your calculator probably doesn't have
an operator's seat with a deadman switch in it.
I. Martins
I believe what you are describing is a mistranslation or, if you want,
a semantic error in the sense that it doesn't mean the same thing that
the original sentence was saying.
As far as I remember, Syntax covers everything related to the building
of a sentence (usage of nouns, pronouns, agreement, etc.) so any
syntax of error is, ipso facto, a grammatical error. The only thing I
can think of is maybe ambivalence. A sentence like "John saw the girl
with the binoculars" can mean:
John saw (the girl with the binoculars) or (John saw the girl) with
<using> the binoculars
Both are grammatically correct but the syntactic analysis of the
sentences differs creating some ambivalence.
However, I think that for most intents and purposes specifying
"grammatical errors" and "syntax errors" is just a repetition.
Then again, I'm not a linguist so I may be wrong.
My USD0.02
Isaac
Andy Lausberg
Excuse me, but I think there's something defective
with the original assignment. As I understand it,
language has three features: semantics (= meaning),
syntax (= grammar, which words fit with which), and
pragmatics (= social relationships, e.g., whether
one does or does not use "sir" when addressing
someone; "pragmatics" sure is a misleading term
of art). That is, "syntax" is just a highfalutin'
word for "grammar". If anyone knows of a nontrivial
distinction between syntax and grammar, do tell.
I'm probably getting too rusty at this for my own good, but:
I think that "this is an book"* (vs. "this is a book") would qualify as a grammatical error, but not as a syntactic error. The position and use of the indefinite article are correct, but the form is not. Likewise, "this is a books"*.
"This chicken is a fine mammal"* qualifies as a semantic error, but not a grammatical or syntactical one.
Syntax specifically has to do with word order/sequence and/or word combination to form larger units such as phrases, clauses and sentences. Grammar, in the linguistic sense, and also in the mundane sense, encompasses a much broader area than mere syntax, although the linguistic and the mundane senses of grammar differ. Grammar also includes morphology, that is, how words are formed.
In reference to your proposed 3-aspect frame for language, I'd say a better description would be 1) semantics 2) grammar (of which syntax is a part) and 3) phonology/phonetics. Pragmatics is actually a sub-field or an adjunct to semantics, as far as I recall.
An explanation of these is: 1) deals with the structure and function of meaning, 3) deals with how language is expressed in communicable form and 2) deals with how 1) and 3) are fused together to form actual instances of language.
Andrew Lausberg, B.A., Grad. Dip.
Korean, Japanese and English Translation
laus...@oceanreach.org
www.oceanasia.com
61-4-6656-9621 (Australia)
Brian Chandler
[but not necessarily in this order]
> What's a little bit unclear to me is the difference between a
> grammatical error and a syntax error. Maybe grammar is a bit broader
> than syntax, and would include things like number and tense. So for
> example:
>
> "I ate one apples."
>
> I thinks that would be a grammatical error.
-- syntax certainly includes things like number agreement. But going
back:
> 2011/8/19 Stephen A. Carter <sca...@hticn.com>:
>
> > That would have been a syntax error, as I understand it. (The original
> > Japanese was something along the lines of 「一定時間放置後のオートオフ」.)
("syntactic") mistranslation. Which brings us back to the orginal
question:
> I came across a proofreading checklist that contained, among other
> things, the following items:
the Chicago Manual of Style, or "proofreading" as in "here's a
translation; try to make it real English". Fundamentally proofreading
is supposed to be looking for printer's errors and author's slips, not
mistranslations.
Brian Chandler
Herman
> word for "grammar". If anyone knows of a nontrivial
> distinction between syntax and grammar, do tell.
While syntax within linguistics is a part of grammar, in the context of
proofreading, there could be problems of poor syntax (i.e. a clumsy,
unnatural, unclear, overly complex, overly long, or otherwise
problematic or inappropriate sentence structure) which are not strictly
speaking grammatical errors. For example, patent claims are supposed to
have a particular syntax, and a deviation from that could be
characterized as a syntax error, even if it is grammatically correct.
Herman Kahn
Steven P. Venti
> However, I think that for most intents and purposes specifying
> "grammatical errors" and "syntax errors" is just a repetition.
I agree and think that the second of the three probably should have been
"usage errors."
But what can you expect when, as Brian Chandler alluded, far too many people
do not maintain a clear distinction in their minds between proofreading and
copy editing.
http://en.wikipedia.org/wiki/Proofreading
http://en.wikipedia.org/wiki/Copy_editing
FWIW
-----------------------------------------------------------------
Steve Venti
spv...@bhk-limited.com
Sittin' by the water, stare into the stream,
Tell me, honey, are there any catfish in your dreams?
--Danny O'Keefe
-----------------------------------------------------------------
Doreen Simmons
FWIW,
Doreen Simmons
Brian Chandler
> Grammar is about the building blocks. Syntax is how they are put together to make a building.
blocks" of language that are not about how they fit together?
I don't know -- there might be a sort of traditional meaning I don't
know about, but you can't get much more traditional than Thomson and
Martinet's "Practical English Grammar", which is definitely about
everything, including syntax. And Pinker in Language Instinct says of
syntax that it is the "component of grammar" that does the fitting
together, presumably leaving things like morphology as other
components. Anyway, it's extremely hard to imagine what an error of
syntax that was not an error of grammar would look like.
Brian Chandler
Steven P. Venti
> Anyway, it's extremely hard to imagine what an error of
> syntax that was not an error of grammar would look like.
Not sure I can agree with that. Many modifiers are clearly grammatically well
formed but syntactically incorrect.
Regrettably, a good example does not immediately come to mind.
FWIW
-----------------------------------------------------------------
Steve Venti
spv...@bhk-limited.com
Catfish ain't expensive, neither is it free,
Some folks crazy 'bout it, others have to leave it be.
--Danny O'Keefe
-----------------------------------------------------------------
Steven P. Venti
> Many modifiers are clearly grammatically well formed but syntactically incorrect.
Arrgh! Many _dangling modifiers_ are clearly grammatically well formed but
syntactically incorrect.
Time to quit while I'm not too far behind.
-----------------------------------------------------------------
Steve Venti
spv...@bhk-limited.com
Oh, you go down to the bank so you can cast your bait,
If you're looking for a catfish, mama, you ain't got long to wait.
Catfish got whiskers and a cute little grin,
But you never can tell where a catfish has been.
--Danny O'Keefe
-----------------------------------------------------------------
Neil Ramsay
> Hmm. Well is this "proofreading", as in the subject of Chapter 3 of
> the Chicago Manual of Style, or "proofreading" as in "here's a
> translation; try to make it real English". Fundamentally proofreading
> is supposed to be looking for printer's errors and author's slips, not
> mistranslations.
The "proofreading" I was referring to was the type described in the
Chicago Manual of Style, not the "here's a
translation; try to make it real English"-type.
Regards,
Neil
Neil Ramsay
have a much clearer idea of the distinction the writer of this
checklist was trying to make. Whether they are suitable things to
check for in the proofreading stage or not, I don't know.
Hope you all have a nice, relaxing weekend.
Neil
> --
> You received this message because you are subscribed to the Honyaku Mailing list.
> To unsubscribe from this group, send email to honyaku+u...@googlegroups.com
> For more options, visit this group at http://groups.google.com/group/honyaku?hl=en?hl=en
>
Doreen Simmons
declensions of the noun and adjective, and the conjugation of the verb.
Then, if you were like me, you move on to learn that Icelandic has more
than 40 declensions (tho' most have minimal differences). But long
before that epiphany I went (3rd year uni) into elementary classical Hebrew and
discovered that, in this world of discourse, a verb wasn't a
conjugational verb as I had always understood it; --it might be, for example,
"informal non-past" -- and I realised, at the age of around 21, that all my
(largely private) study of several Indo-European languages had left me totally unprepared for other
basically different language systems. This revelation was to stand me in good stead when
I took up residence in Singapore in the 1960s and needed to get around in Malay
-- another, totally different, language system. By the time I got to Japan in
the early 1970s I was into my third totally different language base. It didn't help,
of course, that at that time (and to some extent even today) the older Japanese
still believe that they are dealing with uneducated Americans of the Occupation Forces.
Doreen, going back two whole generations here....
>Doreen=E3=80=80Simmons wrote:
>> Grammar is about the building blocks. Syntax is how they are put together=
> to make a building.
>
>How does that work, then? What facts are there about the "building
>blocks" of language that are not about how they fit together?
>
>I don't know -- there might be a sort of traditional meaning I don't
>know about, but you can't get much more traditional than Thomson and
>Martinet's "Practical English Grammar", which is definitely about
>everything, including syntax. And Pinker in Language Instinct says of
>syntax that it is the "component of grammar" that does the fitting
>together, presumably leaving things like morphology as other
>components. Anyway, it's extremely hard to imagine what an error of
>syntax that was not an error of grammar would look like.
>
>Brian Chandler
Brian Chandler
> Thanks to everyone for helping me to get my head round this. I now
> have a much clearer idea of the distinction the writer of this
> checklist was trying to make.
have no idea at all what distinction the original writer was trying to
make, so please explain what you think it is.
Brian Chandler
Neil Ramsay
> make, so please explain what you think it is.
Presumably that, because (as Steve said) some phrases or sentences may
be grammatically well formed but syntactically incorrect, the person
performing this job is required to check for both grammar and syntax
errors. Incidentally, a guy at work pointed out another type of
problem that would presumably not qualify as being a grammar error:
syntactic ambiguity (see the following for an example).
http://www.koncision.com/a-little-syntactic-ambiguity-a-lot-of-time-and-money-wasted/
Regards,
Neil
Benjamin Barrett
The classic example for good syntax with poor semantics is "Colorless green ideas sleep furiously." The button that goes off in your head giving you a warning is your semantics button. That's opposed to something like "head my there are ideas in lots of" where the meaning is all right, but the syntax (grammar) is wrong.
Benjamin Barrett
Seattle, WA
Rene
"Time flies like an arrow, fruit flies like a banana."
Gnaw on that :-)
Rene von Rentzell
Tokyo
Brian Chandler
[me]
> > make, so please explain what you think it is.
>
> Presumably that, because (as Steve said) some phrases or sentences may
> be grammatically well formed but syntactically incorrect, the person
> performing this job is required to check for both grammar and syntax
> errors.
Benjamin Barrett said: "Syntax is what linguists call grammar". (A
delicious ambiguity itself, which can be resolved by putting quotes
around 'syntax' or 'grammar'!) Arguably, 'syntax' can be used to refer
to only "sentence-level" grammar, leaving 'morphology' (or "how words
are made") as a separate part of grammar. But (a) traditional grammar
books certainly cover morphology, though of course they don't call it
that, and (b) the distinction (between syntax and morphology) is a
slightly woolly one at best.
I should say that a previous thread ("Long, complex sentences") makes
it fairly clear that Steve Venti doesn't use (or even accept) the
normal definition of "grammar" used by linguists -- see here:
http://honyaku-archive.org/search/?q=%22long%2C+complex+sentences%22
There is an informal usage of "ungrammatical" to mean "I can't
understand it"; but it looks as though he now calls this "grammatical
but syntactically incorrect". I don't recognise any of this
terminology at all.
To go back to proof-reading, the proof-reader is also supposed to look
out for infelicities, which are things like ambiguous attachments
("For Sale, bath for baby with tin bottom") or larger scale disasters
("Home wanted for bulldog puppy. Eats anything. Very fond of
children.") There is nothing whatever violating grammar/syntax here,
but quite possibly your writer intends to include these in one or
other of the categories. Who knows which?
> http://www.koncision.com/a-little-syntactic-ambiguity-a-lot-of-time-and-money-wasted/
Yes, this is syntactic ambiguity, which is not the same as a syntax
error, and which is rife in natural languages. It's (just) one reason
that MT between E-J is a nightmare. It isn't allowed in programming
languages, because they are required to parse unambiguously, but
happens -- see the "dangling ELSE" of Algol60.
Brian Chandler
Steven P. Venti
> To go back to proof-reading, the proof-reader is also supposed to look
> out for infelicities, which are things like ambiguous attachments
> ("For Sale, bath for baby with tin bottom") or larger scale disasters
> ("Home wanted for bulldog puppy. Eats anything. Very fond of
> children.")
Now I see why we don't understand each other, Brian. You think that
proofreaders do copy editing.
Herman
> out for infelicities, which are things like ambiguous attachments
> ("For Sale, bath for baby with tin bottom") or larger scale disasters
> ("Home wanted for bulldog puppy. Eats anything. Very fond of
> children.") There is nothing whatever violating grammar/syntax here,
> but quite possibly your writer intends to include these in one or
> other of the categories. Who knows which?
>
>> http://www.koncision.com/a-little-syntactic-ambiguity-a-lot-of-time-and-money-wasted/
>
> Yes, this is syntactic ambiguity, which is not the same as a syntax
> error, and which is rife in natural languages. It's (just) one reason
> that MT between E-J is a nightmare. It isn't allowed in programming
> languages, because they are required to parse unambiguously, but
> happens -- see the "dangling ELSE" of Algol60.
>
> Brian Chandler
>
A genuine syntactic ambiguity (as opposed to vagueness or
underdeterminedness) which cannot be resolved from context could in some
cases compromise the communicative function of a text, and in such cases
it would be a 'syntactic error' which a proofreader should correct, even
if the text 'works' grammatically -- just like a syntax checker for a
computer language should flag syntactic ambiguities, even if the
compiler or interpreter would still be able to compile/interpret the
program text as-is.
Herman Kahn