Okay, I'm not quite sure if i'm allowed to post this here. I had a look at the linguistics SE, but it seems that questions there have to be research-level, and this is extremely elementary however I guess it is language related so I'll try my luck. My questions necessitates reading this passage:

... (11)
A or B
Not A
Therefore B

These schematic representations of arguments are called argument schemata. The letters A and B stand for arbitrary sentences. Filling in actual sentences for them we obtain an actual argument. Any such substitution into schema (11) results in a valid argument, which is why (11) is said to be a valid argument schema.

The form we said that could be represented by (11) is more than just a syntactic construction. The first premise is not just two sentence linked by a conjunction, for it is also important what conjunction we are dealing with. A different argument schema is obtained if the conjunction 'or' in (11) is replaced by another conjunction say, 'if'.

My questions are:

  1. What is meant by a 'syntactic construction'?
  2. Why is the argument schemata represented by (11) "more than just a syntactic construction"?

As syntax is essentially the structure of a sentence, I believe a syntactic construction is merely a schematic representation of the structure of a sentence as in (11). The confusion I seem to be experiencing stems from when the author says "The form we said that could be represented by (11) is more than just a syntactic construction". What makes it more than just a syntactic construction? I then believe he goes on to offer a reason: "for it is also important what conjunction we are dealing with"; surely the conjunction we are dealing with would be important in a syntactic construction too?

Could someone help me understand this please. Thanks.


(11) is an argument schemata because it uses what are called metalanguage variables 'A' and 'B'. These are variables for which you can substitute sentences of your object language (in your case, English). Let A = 'dogs are black' and B = 'cats are happy' and you get:

Dogs are black or cats are happy

Dogs aren't black.

Therefore, cats are happy.

Now let me explain what is meant by 'syntactic construction'.

A syntactic construction just specifies types of syntactic categories. It does not use particular words of the language. This is true in linguistics and whoever wrote your text is also applying the same principle to logic (although many logicians don't conform to this usage). In (elementary) logic, the syntactic categories are sentence, conjunction (also called a two-place connective), and one-place connectives (for example, negation).

Each of these categories has particular words (or strings of words) that belong to it. For example:

  • sentences: {dogs are black, cats are happy, ...}

  • conjunctions: {and, or, if...then}

  • one-place connectives: {not}

A syntactic construction would use only the category names and not the specific words that belong to those categories. For example:

sentence conjunction sentence

negation sentence


Notice that I have not used any particular sentences or conjunctions. But (11) does use a particular word belonging to the category conjunction, namely 'or'. In that sense, it is more than a syntactic construction.

  • If anyone used the phrase "purely syntactic construction", it might very well mean what you say it means. But the question does not ask about constructions that are "purely syntactic". No "purely" is there, so there is no rationale for excluding specific morphemes from being part of a syntactic construction. – Greg Lee Jun 7 '16 at 14:46
  • @Greg Lee, I thought that my use of 'purely' might be misleading. I'm going to edit it out. It doesn't change a thing... – GrimGrom Jun 7 '16 at 14:58
  • @Greg Lee, you don't use specific morphemes in describing the syntax of logic just like you don't use them in linguistics. Linguists would give the syntactic structure of 'The dog runs' as [NP VP] (that is, noun phrase + verb phrase), or, with more detail [[DET][N]][[V]] (that is, determiner + noun + verb) No specific morphemes or lexical items are mentioned. Isn't this right? – GrimGrom Jun 7 '16 at 15:01
  • So now you've stuck in a gratuitous "merely" to replace the gratuitous "purely". That is not an improvement. – Greg Lee Jun 7 '16 at 15:04
  • 1
    What you say about describing the syntax of logic is just not true, except that a logician would of course not use the term "morpheme", but would refer to specific sentence connectives or operators. Just look at a definition of wff in any logic text to see. – Greg Lee Jun 7 '16 at 15:09

No, it cannot be understood. It's nonsense, so please relax. The author of whatever you're quoting says of his schema (11), the "first premise is not just two sentence[s] linked by a conjunction, for it is also important what conjunction we are dealing with." But look at the schema (11) -- it has "or", not any general conjunction or any other conjunction. Just "or". Right there it is, in black and white.

So if all that could prevent (11) from being a "syntactic schema" is that no conjunction other than "or" could connect A and B, then since "or" is required as the specific conjunction connecting A and B, then (11) is, after all, a "syntactic schema".

Maybe this is from a handout that was not adequately proof read?


Syntax is simply the structure of a sentence. So syntactic construction simply means how you construct sentences. In writing it is helpful to vary your syntax.

Different languages have different syntactical demands. For example, in German the verb always goes second. In English, the normal syntax would be to have a subject/verb/indirect object/direct object.


I don't understand using a verbose phrase like "syntactical construction", when "syntax" would work just as well.

Source: Yahoo! Answers

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.