The grapes aren't sour. Prove it!

You'd think that taste might be an area in which the personal escapes public criticism, but you would be wrong.

Here is Hume on the bad critic:

He must conclude, upon the whole, that the fault lies in himself, and that he wants the delicacy, which is requisite to make him sensible of every beauty and every blemish, in any composition or discourse.

In other words, as shadow might say, such a critic has unreliable taste buds. But how, Hume asks, can we (who?) educate this person? How are we to convince him (and of course, it's always a man, being the 18th century and all) that he needs to revise his opinion of a certain dish?

Success will come:

. . . when we show him an avowed principle of art; when we illustrate this principle by examples, whose operation, from his own particular taste, he acknowledges to be conformable to the principle; when we prove, that the same principle may be applied to the present case, where he did not perceive or feel its influence[.]

Let's go over that again. To succeed at convincing the bad critic of his badness you may take the following steps:
1) Show a principle of taste.
2) Demonstrate that principle by agreed upon examples.
3) Demonstrate that the principle applies to the current object.

So a Hume-an food conversation might go something like this. You have to look pretty carefully to see the operation of the three steps, though, since they often occur all at the same time:

A: I don't like tofu.
B: But smooth-textured and creamy foods are tasty. For example, you like ice-cream and soft cheeses.
A: But it's bland.
B: First of all, tofu has its own bean taste. Secondly, tofu is often eaten with other things. It's only "bland" because it is not a pre-salted product. There are plenty of other unsalted products that are "bland" until seasoned. Stock, for example, or rice.

This is a pretty logical, and verbal, way to go about silencing the bad critic. But in reality, bad critics are often very resilient to persuasion of this type. Far more effective, I have found, is the following method.

A: I don't like broccoli.
*Chef prepares a broccoli dish in which it is cooked right and beautifully presented*
A: *Eating* Oh, this is good.

So then we have two models of silencing the bad critic. The first model is Hume-an persuasion. The second model is a sort of proof by counterexample. Silencing the critic, as it were, by putting food in his mouth.

The chief limitation of the first method is that it rarely works.

The chief limitation of the second method is that it does not do so well for persuading people that they do not or should not like a certain food.

Another limit on the second method is that it's not so great for persuading people that they should like a specific dish, at least, not directly, unless the first time they tasted the dish, it was poorly prepared. However, one can always combine the two methods to overcome this second limit.

Imagine that a person claims not to like the Penne Puttanesca at restaurant X. First one might try to find out what exactly the person does not like about it. Whether, for example, he dislikes that puttanesca, or all puttanescas, or perhaps all pennes. Then, depending on what you found out, you might try to present the person with dishes that prove the critic wrong. In an especially intractable case, you might need to introduce a whole series of sauces and starches, working your way into the neighborhood of Penne Puttanesca.

The second method is inherently kind. To show the bad critic his error, you must feed him.

But of course, the second method is inherently violative. To show the bad critic his error, you must feed him food that there is a risk he will dislike, possibly violently so.

This kind violation is exactly the kind of intimacy elided by Hume's persuasion.

So why talk about food at all?

It is a happy occurrence that the purpose and motivation for conversation about food may coincide.

What is the motivation? As Scarry says, "beauty brings copies of itself into being". We are motivated by the food itself to talk about the food, "as when the seen face incites an ache of longing in the hand, and the hand then presses pencil to paper".

Which means when we talk about food, we are driven to replicate its beauty in words. As I have noted above, the mere logic of an argument proves inadequate to persuade a bad critic of a dish's deliciousness, or beauty. An argument, then, can only succeed if it relies on its own beauty, the same way another dish "demonstrates" a principle of taste.

Witness the outrage at (others) being "deceived" by a critic such as Mr. Gold, whose writing is so carefully crafted. A food review is a form of argument. When an argument has a beauty of its own, it can feel as intimate, as kind or as violative, as a meal.