This may not be critical; it's the first thing that comes to mind.

The kind of ideas that an idea theory of meaning requires seem to be created after the fact and therefore don't have explanatory value. Concentrating specifically on a mental grouping with necessary and sufficient conditions. Let's take the example 'car'. By your very hypothesis there is no separation between the idea of a car and all the ways we use the word car.
Why is it that we have the idea car? It's obvious that when we learnt the idea we were not initially told the necessary and sufficient conditions for something to fit with that idea. Why are a 2 seated sports car and a 5 seated sedan both considered cars but a van isn't. The answer is that our language treats them differently rather than them being natural ideas we had which we tried to explain with language.

So the idea of a car is derivative on all the way we use the word in language not the cause of us using those words. If I show you some pictures and ask 'is this one a car?', 'is this one a car'. To find your answers you think to yourself 'would I use the word car to describe this'. You don't run through a mental checklist of conditions necessary and sufficient conditions.

Let's say that we run through the above exercise where I show pictures. Before the exercise I could ask you to list the necessary and sufficient conditions for an object to be a chair. Inevitably there will come some pictures which you would use the word chair to describe but don't fit the list of conditions. There is a question of how you react to these examples, will you argue that we have been using the word chair incorrectly as the object does not fit your conditions or will you change the list of conditions to make them inclusive the object pictured.

To hold your view you must state that the former is impossible because we can't use the word 'chair' if it doesn't fit our idea specifically because the word is used to express the idea. The latter is also impossible because then we need to say that we don't have a complete idea of a chair with all its necessary and sufficient conditions when we use the word. It's admitting that we can only work out the necessary and sufficient conditions from inspecting the way we use the word.

In the end what this comes down to is questioning whether this idea exists. I can't find it in the phenomenology of my use of language. If we can only get to the ideas by inspecting our use of language how can they explain our understanding of language. Finally how do I understand ideas? If I have some kind of direct way of understanding them what is it about them that allows for it but doesn't allow for a direct way of understanding sentences without ideas as an intermediary?

Some more simple questions. What is the idea behind the meaning of 'hello'? What is the idea of '5723'?

Why is it that we have the idea car? — shmik

We learn ostensively, learning by examples then constructing a rule (necessary and sufficient conditions) that admits of those examples.

To find your answers you think to yourself 'would I use the word car to describe this'. You don't run through a mental checklist of conditions necessary and sufficient conditions. — shmik

Not consciously maybe, but if knowing a meaning is knowing the rule, then thinking 'would I use the word "car" to describe this' IS comparing your thought to the rule (i.e, a checklist of necessary and sufficient conditions)

Inevitably there will come some pictures which you would use the word chair to describe but don't fit the list of conditions. — shmik

Why would this be so?

Finally how do I understand ideas? If I have some kind of direct way of understanding them what is it about them that allows for it but doesn't allow for a direct way of understanding sentences without ideas as an intermediary? — shmik

I'm confused by this. Can you explain what you mean 'sentences without ideas as an intermediary'?

Some more simple questions. What is the idea behind the meaning of 'hello'? What is the idea of '5723'? — shmik

Hello just expresses the idea of a greeting. 5723 expresses the idea of an integer after 5722. I'm assuming these answers are not what you are after, but I wonder why not?

I don't really get your position. I can't see how you are arguing for an ideational theory of meaning.

We learn ostensively, learning by examples then constructing a rule (necessary and sufficient conditions) that admits of those examples. — invizzy

Not consciously maybe, but if knowing a meaning is knowing the rule, then thinking 'would I use the word "car" to describe this' IS comparing your thought to the rule (i.e, a checklist of necessary and sufficient conditions) — invizzy

If knowing a meaning is knowing a rule then where do ideas 'private and independent of language' come in.

It would help if you made an example of how ideas connect with rules which connect with language on both the part of a speaker and listener. i.e the idea in the Jim's head is converted into language by ... which is then... which is then converted into an into an idea in Bob's head for him to understand it. Or something like that.

OK, like I say, meaning is an idea. What is an idea though? Simply a mental grouping, a type of thought articulated by a rule/necessary and sufficient conditions.

So the speaker has a specific thought, then an idea/meaning/mental grouping is chosen which conveys that thought (i.e. the thought fits the conditions), then the associated word or expression for that meaning is uttered, then the hearer receives the meaning (i.e. the mental grouping), and you've just had successful communication.

For the life of me and I can't see the flaws of this, it is just a simple elegant theory.

So, if you have a thought that you want to greet someone you can choose the word 'hello' which has an appropriate meaning/grouping for that thought.

Isn't this a lot simpler than Propositional Theories of Meaning?

So the speaker has a specific thought, then an idea/meaning/mental grouping is chosen which conveys that thought (i.e. the thought fits the conditions), then the associated word or expression for that meaning is uttered, then the hearer receives the meaning (i.e. the mental grouping), and you've just had successful communication. — "invizzy

The flaw is in the relationship of language to ideas. Here the meaning of the (unspoken) idea/meaning/mental grouping is the same as the associated word or expression. In meaning, there is no separation between the unspoken idea and what is meant by words said to communicate it. Ideas, in terms of meaning, are clearly not independent of language. More critically, the specific thought speaker has, if it is given by idea/meaning/mental grouping is not distinct from that idea/meaning/mental grouping. Ideational theories use an incoherent relationship between ideas and meaning. They make a distinction which is not there.

The distinctions used under ideational theories are more akin to distinction of states of the world related to thinking and communicating. I might, for example, have the thought I need to explain something (initial thought), think and work what meaning I need to say to communicate it to others (idea/meaning/mental grouping which, critically, does not convey my initial thought; I am not telling people I need to tell them something when I explain the issue), then move on to speaking the words which represent the idea/meaning/mental grouping I had.

We learn ostensively, learning by examples then constructing a rule (necessary and sufficient conditions) that admits of those examples. — invizzy

How does one ostensively learn the meaning of "obligation"?

Ideational theories use an incoherent relationship between ideas and meaning. They make a distinction which is not there. — TheWillowOfDarkness

I don't think I am making a distinction in my theory though. Can you see that?

The distinctions used under ideational theories are more akin to distinction of states of the world related to thinking and communicating. I might, for example, have the thought I need to explain something (initial thought), think and work what meaning I need to say to communicate it to others (idea/meaning/mental grouping which, critically, does not convey my initial thought; I am not telling people I need to tell them something when I explain the issue), then move on to speaking the words which represent the idea/meaning/mental grouping I had. — TheWillowOfDarkness

Can you elaborate on this? What are you saying here? You seem to be agreeing with a version of an ideational theory, right? But then why doesn't the idea convey the original thought, as with my theory?

How does one ostensively learn the meaning of "obligation"? — Michael

By listening to how people use the word obligation, determining the similarities in the things called obligations and using that rule when we ourselves use the world.

But obligations aren't the sort of things that can be shown, like cars or cups. I can point out a knife and a person, but I can't point out an obligation not to use the knife to stab the person. When you say "you ought not stab that person with that knife", how am I to know what you mean by "ought not"?

By listening to examples of things that speakers say we 'ought not' do (like stabbing people with knives) and inferring based on knowledge of speakers values, and facts about stabbing people, and the situation at hand, the sort of things the speaker would be saying.

I don't think that counts as an ostensive definition. An ostensive definition is where you point at something and say "that's a chair", "that's the colour red", "that's a person", and so on. So to learn the meaning of "obligation" ostensively one would have to point to a thing and say "that's an obligation". That's different to how you've explained it above.

Perhaps you are technically correct. Is there another word for learning meanings in the way I suggest?