Burgess Laughlin

What is grammar?

15 posts in this topic

I need to start by clarifying what "grammar" is. Is it an art or is it a science? My understanding is that grammar is a science, in the sense that it is, for some individuals, a systematic study of something (here, the use of words to form sentences). It is an art too, in the sense that one can systematically apply the knowledge developed by the science, but the application requires judgment, that is, consideration of one's purpose and the context, and so the rules are flexible.

If grammar is both a science and an art, then grammar is both descriptive and prescriptive. That is, grammar describes what objective sentences look like, and prescribes rules that should guide us in speaking, writing, and thinking. Thus, if my idea of "grammar" is correct, it studies things as they are used (in their proper form, at least) and tells us how they ought to be used.

If grammarians focus on sentences, why? Perhaps it is because a sentence is the basic unit of thought. A sentence says something about something. An example is: "This computer is a laptop." I am saying something ("is a laptop"), as the predicate of the sentence, about the subject ("This computer") of the sentence.

Last, my dictionary says there are two main branches of grammar:

1. Morphology (the "study of forms") examines issues such as how words change to show different meanings. Perhaps an example is this inflection (a systematic change in endings to show changes in usage and meaning): He, his, him.

The three words refer to the same thing, a male, but the systematic change in the endings (inflection) shows that that male is being referred to in different ways:

- "He is the owner of the business." (Shows the subject)

- "The book is his." (Shows possession)

- "I gave him the book." (Shows an object, actually here an indirect object: "to him")

A more common form of change in a word to show a change in meaning is changing singular to plural: From child to children, boy to boys, berry to berries.

2. Syntax examines the ways words can be arranged to logically convey meaning. For example, in English we generally place a subject first, then the verb: "The boy runs" rather than "Runs boy the."

The above are the basic points I want to offer for criticism before talking about particular issues. Corrections? Additions?

Are there other basic concepts that beginners should know before talking about particular rules?

Share this post


Link to post
Share on other sites
I need to start by clarifying what "grammar" is. Is it an art or is it a science?
If it is a science, what is the object of study, and what are its methods? Physics is a science, which studies the nature of existence, and its methods and goals are very different from an art like sculpture. What are other examples of disciplines that are both art and science -- how do people define those disciplines to avoid equivocation? What do you mean by grammar? What are you trying to distinguish it from?

Share this post


Link to post
Share on other sites

I wouldnt say that grammar is any more of an art or science than 'spelling'. They play pretty much identical roles - standardised grammar and spelling are linguistic conventions adopted by a community in order to make communication easier. If you want to actually study a language then the grammar probably provides more interesting topics to discuss and publish papers on, but I dont see them as being fundamentally different.

Share this post


Link to post
Share on other sites

If grammarians focus on sentences, why? Perhaps it is because a sentence is the basic unit of thought. A sentence says something about something. An example is: "This computer is a laptop." I am saying something ("is a laptop"), as the predicate of the sentence, about the subject ("This computer") of the sentence.

Is this true in all human languages, or just Western ones? As far as I know, subject/predicate dualism is a lot more pronounced in Indo-European languages, and hence Western philosophy, than it is elsewhere (Chinese for instance). I dont claim to know much about this, so someone else might want to correct me if I'm mistaken.

Share this post


Link to post
Share on other sites
Is this true in all human languages, or just Western ones? As far as I know, subject/predicate dualism is a lot more pronounced in Indo-European languages, and hence Western philosophy, than it is elsewhere (Chinese for instance). I dont claim to know much about this, so someone else might want to correct me if I'm mistaken.

Hi. That is generally true that Indo-European languages lean towards the subject/predicate distincition more so than others.

In many languages, the division can be more accurately thought of as that of topic/comment rather than subject/predicate. The topic is what the speaker wants to talk about and is asserted first. The comment about the topic follows. This division is also more prevalent in speech, regardless of which language. For example one could utter: 'Bears ... beautiful animals' or 'Running around... my son loves doing that'. (The first bit is the topic.) Obviously there is some commonality between subject/predicate and topic/comment, especially if a language provides a subject/predicate mechanism. Another way to look at it is that the idea of what a sentence is in non-subject/predicate languages is a bit looser or not as developed as a tight unit.

Share this post


Link to post
Share on other sites

David, thank you for the questions. I will try to answer them to the extent that I, as a novice student of English grammar, can. Perhaps my answers will provide a target for correction, so that the purpose of this thread -- which is to provide preliminary basics for those who want to learn better English grammar -- can be fulfilled.

I have numbered your questions and one comment into separate, numbered points, to make keeping track of them easier.

[1]If it is a science, what is the object of study, and [...]

[2] [W]hat are its methods?

[3]Physics is a science, which studies the nature of existence, and its methods and goals are very different from an art like sculpture.

[4] What are other examples of disciplines that are both art and science -- how do people define those disciplines to avoid equivocation?

[5] What do you mean by grammar?

[6] What are you trying to distinguish it from?

[1] As I tried to indicate in my first post, an English grammarian's object of study is the sentence.

[2] An English grammarian's methods would include, of course, the methods of philosophy, which are methods available to everyone, everywhere, and at all times: sense-perception, thinking (in sentences, including questions), and logic (to test the validity of the thinking).

Perhaps you intended to ask this question: If grammar is a specialized science, then what special methods would English grammarians employ? I don't know, beyond such approaches as identifying words by part of speech and sentence-diagramming. Can you suggest others?

[3] Is the "nature of existence" the special object of study by physicists? I doubt that. That would be the object of meta-physicists. I would think that the physical world would be the object of study for physicists, perhaps from some special, delimiting viewpoint to distinguish it from studies of physical life-forms, for example. I don't know. That is outside my area. What would you suggest as an alternative?

Yes, physics has methods and goals which "are very different from an art like sculpture." Physics is a science, and sculpture is a fine art (as distinct from an "art" (techne) in the sense of the "art of steel making"). What is your point?

[4] Logic is both an art (or should be, for all of us) and a science (for a few people, called logicians, whether they are full-time or part-time or devoted to it only as one aspect of their work, as Ayn Rand was in identifying fallacies such as the stolen concept).

If a single term, such as "grammar," appears, the reader can determine whether it refers to the science or the art by examining the context in which the term is used. If that does not make the issue clear, then a listener -- though not a reader -- can ask questions. The writer can make clear his meaning, even if he uses a single term, either by context generally or by adding a qualifying phrase such as the art of grammar or the science of grammar.

Another example of art and science, though not widely recognized, is etiquette. Someone looks at situations and develops rules; most others apply -- or don't apply -- those rules to social situations for the purpose of facilitating trade.

[5] Didn't I answer this in my first post? That is my best answer to this question. What would you suggest as a definition?

[6] I was trying to identify its essential nature as distinct from other science/arts. Do you believe that is too narrow or too broad for the purpose of preparing novices in English grammar for solving their particular grammar problems?

Share this post


Link to post
Share on other sites
Is this true in all human languages, or just Western ones? [...]

How will your questions and comments help fulfill the purpose of the grammar school -- which is to help Objectivists in this forum who are having grammar problems learn better English grammar in their writing?

Share this post


Link to post
Share on other sites

I'll elaborate on a few of my questions and your answers, to focus on a particular metaphysical issue about what grammars.

[1] As I tried to indicate in my first post, an English grammarian's object of study is the sentence.
That's very succinct, and I could agree with it, if you mean what I would mean. Let's find out. An English grammarian's object of study is the nature of any and all sentences that are in the English language. You would be interested in things that aren't sentences or things that aren't "in English" only insofar as it elucidates English sentences and leads to criteria for differentiation. Here's the ontological problem (I think it's a small one, but it's important to keep it in mind): there has to be an objective way of determining if something is a sentence in English, or isn't, if you want to consider grammar from a scientific perspective. Traditional approaches to grammar, which take the "art" perspective and not the "science" perspective, solve the problem easily -- by checking against a set of known and fixed rules, which have been published in various books and papers over the past 200 years (perhaps excluding the past 40 years -- I don't have much opinion on such works since the Decline of Grammar).

The other approach, the "science" approach, faces the problem of independently justifying particular decisions, for example is this a sentence of English: "That's the person I was talking with." -- some people feel that it is not. If you are trying to find the defining characteristics of word sequences that are sentences in English vs. are not -- e.g. "*That's the person with I was talking." or "Du må brukke engelsk med oss" -- you need a different objective basis for deciding "what's in" vs. "what's not in". The perspective that I adopt, considering grammar as a science, is a psychological one: a grammar is a system of cognitive principles that actually exist in the minds of people who speak languages, which is causally connected to the way that sentences are performed and understood by the particular speaker of the language. As such, a grammar itself can't be directly inspected with current technology, the way a stomach can be inspected. However, we can create models of grammars, theories of what to the best of our understanding is actually going on in the mind. Thus from my POV, the study of grammar is part of the study of individual cognition, which cannot be divorced from the facts of the person or persons who you're talking about.

I think these two perspectives on what a grammar is can be very simply explained on the basis of different purposes. If I understand your goal correctly (and that is a big if), the scientific perspective on grammar would really be more of a hindrance than a help. I presume that your goal relates to good writing -- if so, you don't need the fine-grained details of the science of grammar, any more than a microbiologist needs string theory to do their work.

[4] Logic is both an art (or should be, for all of us) and a science (for a few people, called logicians, whether they are full-time or part-time or devoted to it only as one aspect of their work, as Ayn Rand was in identifying fallacies such as the stolen concept).
Nice: I was thinking that this might be an example. So to state the difference more generally, science demands a more explicit and probably abstract account of the nature of the object being studied, with the purpose in mind of developing an objective theory of the object. And an art would be more "intuitive" (I don't mean that in a derogatory manner). Grammar -- and probably most empirical areas -- can be science or art, and the question for you to consider is, how much of the scientific paraphenalia of the study of grammar do you require for your purposes?

Share this post


Link to post
Share on other sites
I need to start by clarifying what "grammar" is. Is it an art or is it a science?

I would call it a normative science -- a systematic study of an aspect of reality resulting in principles (science) that prescribes what a person ought to do (that's the normative part).

Other examples of normative sciences are ethics and epistemology.

Betsy @ OCON

Share this post


Link to post
Share on other sites
Hi. That is generally true that Indo-European languages lean towards the subject/predicate distincition more so than others.

In many languages, the division can be more accurately thought of as that of topic/comment rather than subject/predicate.

I'm not sure just how relevant this comment of mine will be, but it seems related in some important way. I think I am not at all sure about your comment that the subject/predicate distinction might not be as accurate in other languages. I think it is more the case that grammarians from other languages have chosen a different way of describing their language (i.e., they have created different concepts to make sense of the grammar), rather than that our ways/concepts don't accurately also get at their language. I'll relate a story that leads me to believe this.

I recently took an advanced linguistics course, and many students (including the teacher) wanted to say that our western way of viewing grammar was only applicable to western languages. They emphasized how other non indo-european worked very differently, and my "western" science of grammar was not applicable to them. I found this hard to imagine, but it was also true that i had only studied indo-european languages thus far (mainly Greek, with some German, Latin and French). So last summer I decided to begin studying Arabic. Not only was I very interested in finding out whether their comments were true, but I also wanted to dig into Medieval Arabic philosophy (as a sub-discipline to my studies of Greek philosophy--since the Arabs wrote many great commentaries on it). The text we used (al-kitaab) went about describing the grammar of Arabic in the manner that Arabic grammarians so described it; and this was FAR different in many ways from the Western way. But what I found, understanding grammar quite well from my study of Greek, was that Arabic sentences could usually be described just as easily using the western concepts. In fact, I think that many students would have learned things much more quickly had they been able to use the western concepts since these were more familiar to them--we ended up not only having to learn Arabic, but having to learn a new system of grammar.

One of the problems is that people often think that words are a certain part of speech outside the context of a sentence--this is simply untrue. A word can only be said absolutely to be this part of speech or the other in the context of a sentence; though in English it is fairly true that words have only one (or a very few) part of speech in any context, thus people think context isn't important. But in Arabic words are regularly used as different parts of speech--thus they are not categorized as being any one particular part of speech. This leads people to think that the languages cannot be described with the same concepts. But if one understands context is essential, then western concepts quite nicely can be used in Arabic. A simple example in English might be "drive." There is really no part of speech for this word without some given use of it in a sentence. It could be a noun: (for golfers)"That was a great drive." Or it could be a verb: "I drive to school everyday," or even (after having robbed a bank) "Drive!" Thus, if one drops the important context which gives our concepts force, one ends up thinking our concepts are impotent for describing other languages.

Share this post


Link to post
Share on other sites
I'm not sure just how relevant this comment of mine will be, but it seems related in some important way.  I think I am not at all sure about your comment that the subject/predicate distinction might not be as accurate in other languages.  I think it is more the case that grammarians from other languages have chosen a different way of describing their language (i.e., they have created different concepts to make sense of the grammar), rather than that our ways/concepts don't accurately also get at their language.  I'll relate a story that leads me to believe this. 

I recently took an advanced linguistics course, and many students (including the teacher) wanted to say that our western way of viewing grammar was only applicable to western languages.  They emphasized how other non indo-european worked very differently, and my "western" science of grammar was not applicable to them.  I found this hard to imagine, but it was also true that i had only studied indo-european languages thus far (mainly Greek, with some German, Latin and French).  So last summer I decided to begin studying Arabic.  Not only was I very interested in finding out whether their comments were true, but I also wanted to dig into Medieval Arabic philosophy (as a sub-discipline to my studies of Greek philosophy--since the Arabs wrote many great commentaries on it).  The text we used (al-kitaab) went about describing the grammar of Arabic in the manner that Arabic grammarians so described it; and this was FAR different in many ways from the Western way.  But what I found, understanding grammar quite well from my study of Greek, was that Arabic sentences could usually be described just as easily using the western concepts.  In fact, I think that many students would have learned things much more quickly had they been able to use the western concepts since these were more familiar to them--we ended up not only having to learn Arabic, but having to learn a new system of grammar. 

One of the problems is that people often think that words are a certain part of speech outside the context of a sentence--this is simply untrue.  A word can only be said absolutely to be this part of speech or the other in the context of a sentence; though in English it is fairly true that words have only one (or a very few) part of speech in any context, thus people think context isn't important.  But in Arabic words are regularly used as different parts of speech--thus they are not categorized as being any one particular part of speech.  This leads people to think that the languages cannot be described with the same concepts.  But if one understands context is essential, then western concepts quite nicely can be used in Arabic.  A simple example in English might be "drive."  There is really no part of speech for this word without some given use of it in a sentence.  It could be a noun: (for golfers)"That was a great drive."  Or it could be a verb: "I drive to school everyday," or even (after having robbed a bank) "Drive!"  Thus, if one drops the important context which gives our concepts force, one ends up thinking our concepts are impotent for describing other languages.

It depends somewhat on what you mean by subject/predicate. Strictly speaking, subject should equate to allowing only certain parts of speech to act as subject, eg nouns, noun phrases and predicate is headed by a verb or a verbal phrase. In the topic/comment type languages, there are far fewer restrictions or rules.

Arabic (and the Semitic languages in general) has been influenced by thinkers who were exposed to the explicit idea of subject/predicate and are fairly 'Western' as languages go, so I am not surprised that there is a tremendous amount of evidence of subject/predicate. Also, a language that has been studied and has a literary and intellectual history also helps to shape it so that it works much more clearly and precisely.

I have worked in academic linguistics for some time. My research does seem to indicate that as a culture becomes more sophisticated, you do find a tendency for the language to show evidence of development of a stricter sense of subject/predicate regardless of exposure to Western languages. For example, Mandarin which is the Chinese dialect that has been developed the most as the dominant language of literature and learning shows more evidence of having strict subject/predicate rules than do other dialects which are largely spoken. If you trace through its history, you can literally see the development.

Whilst there are many problems with academic linguistics, these sorts of problems are very real. During fieldwork, when I encounter a language that has not really been studied, one of the first tasks is to work out the conceptual units, eg parts of speech, how propositions are made etc as I cannot consult a grammar book for help. It is not automatic that all propositions are represented by the unit 'sentence'. It is also not clear that every word fits neatly into a readily identifiable part of speech.

Share this post


Link to post
Share on other sites

Arabic (and the Semitic languages in general) has been influenced by thinkers who were exposed to the explicit idea of subject/predicate and are fairly 'Western' as languages go, so I am not surprised that there is a tremendous amount of evidence of subject/predicate. Also, a language that has been studied and has a literary and intellectual history also helps to shape it so that it works much more clearly and precisely.

Are you saying that as Arabic became exposed to, say Greek grammar, the language actually tended to develop into a language more in line with these concepts? That's amazing (not in the sense that I don't believe you, just that I didn't know that could happen).

I have worked in academic linguistics for some time. My research does seem to indicate that as a culture becomes more sophisticated, you do find a tendency for the language to show evidence of development of a stricter sense of subject/predicate regardless of exposure to Western languages. For example, Mandarin which is the Chinese dialect that has been developed the most as the dominant language of literature and learning shows more evidence of having strict subject/predicate rules than do other dialects which are largely spoken. If you trace through its history, you can literally see the development.

I would have thought something like that would be the case, since I think there is something about the subject/predicate distinction that accords with the way the world is divided up by the human mind.

Whilst there are many problems with academic linguistics, these sorts of problems are very real. During fieldwork, when I encounter a language that has not really been studied, one of the first tasks is to work out the conceptual units, eg parts of speech, how propositions are made etc as I cannot consult a grammar book for help. It is not automatic that all propositions are represented by the unit 'sentence'.

Would it be possible for you to explain what you mean by this, since I just don't believe it (as I shouldn't on authority--though of course no disrespect is meant). This sounds like the sorts of claims I heard about Arabic which I found to be (so far at least) untrue.

It is also not clear that every word fits neatly into a readily  identifiable part of speech.

I admitted as much in my post. But what I don't believe is that in the context of a sentence (how there could be a meaningful assertion without a sentence I'd also like to hear) there aren't identifiable parts of speech. And this isn't to say that there might not be more or fewer parts of speech than in English--the example I am of course familiar with is the particle in Greek.

Share this post


Link to post
Share on other sites
It depends somewhat on what you mean by subject/predicate.
I've never been persuaded that this is a valid distinction in grammar. Subject is clearly a syntactic property, not a semantic one, and I just don't see the argument that the VP is a coherent semantic unit. Why, for example, not divide the sentence into subject-and-verb, plus "everything else"? Or, to make the cut that I actually prefer, the verb, and its arguments. As for the VP being a syntactic unit, I haven't been terribly impressed by the arguments in general (certainly not for VS languages like Celtic). Given the VSO inclinations of classical Arabic, I'd be interested to see what evidence there is for a concept "predicate" in that language as well.

Share this post


Link to post
Share on other sites
I admitted as much in my post.  But what I don't believe is that in the context of a sentence (how there could be a meaningful assertion without a sentence I'd also like to hear) there aren't identifiable parts of speech.
One of the key words here is "identifiable". If you import a priori concepts into the analysis of a language, i.e. if you hold some version of nativism which assumes that certain aspects of language analysis are given in advance (hard-wired), such as "noun", "verb", "adjective", "adverb", "quantifier" and so on, then you presumably also have some kind of schema of universal category identification which I suppose would be semantically based, so that you don't classify an adverb as a noun. If you have to use just language-internal evidence for creating categories, then the situation will arise that there may not be particularly good evidence that shows what category covers the word in question. So-called particles are one example, and "ideophones" (rampant in certain groups of African languages inter alii, and sort of similar to English words like "squish") are notorious in resisting a definitive parts of speech classification. I am more and more persuaded that while there is validity to the general categories like "adjective", "verb" and "noun", they are not all universally valid at least from an inductive POV, and there are annoying categories like ideophone which cause problems for the usual limited-number perspective on parts of speech.

However, this doesn't have anything to do with English as it bears on writing style, which is the focus of this subforum. Fascinating stuff, nonetheless.

Share this post


Link to post
Share on other sites