Tomatoes are vegetables

Courtesy of Jack Brounstein at The Audhumlan Conspiracy, a story from Reason magazine's blog about a bill designating a variety of tomato as the official state vegetable of New Jersey. Botanists and biologists will no doubt sneer, but SC finds this story strangely compelling.

Let's start by laying out the "scientific" definitions of fruits and vegetables. The catch here is that there really isn't such a definition for vegetables. But we'll deal with that in a second. First, fruit as defined by the Columbia Electronic Encyclopedia:

fruit, matured ovary of the pistil of a flower, containing the seed. After the egg nucleus, or ovum, has been fertilized (see fertilization) and the embryo plantlet begins to form, the surrounding ovule (see pistil) develops into a seed and the ovary wall (pericarp) around the ovule becomes the fruit.

Spend a moment to peruse the "types of fruits" link as well. They're not all exactly prototypical -- peas (when in pods) and beans fall into this classification, as do parts of the carrot plant that you aren't actually likely to find in the grocery store. Bananas are an interesting case -- the inside is a fruit on the technical definition given above, but the skin is actually tissue from the stem of the plant. For more detail on that, see here.

So then there are vegetables. Definitions vary considerably; the UC Davis Vegetable Research and Information Center defines vegetables as "the edible portion of a plant", which would seem to subsume fruits as well, although they then go on to classify vegetables by different plant parts other than ovaries; leaves, stems, roots, etc. A horticulturist observes that, from the standpoint of his profession (and as opposed to botanists), fruits are generally perennial, woody plants, while vegetables are annuals with soft stems. These aren't perfect definitions (watermelon is a vegetable on this reading of things), but as he points out, horticulture isn't an exact science.

Thus, from a purely scientific standpoint, tomatoes are fruits, and there are no such things as vegetables. But this is a plainly counterintuitive result to a typical speaker of English, who knows that he can walk into a grocery store, ask for the location of the vegetable aisle, and get a useful result. Our horticulturist friend points out that:

Tomatoes fit the vegetable category. They are planted every year. We use them in salads and, well, vegetable dishes during the main meal. I haven't had a tomato cake or pie and frankly, don't care if I ever do.

So, that's the foundation. Botanically, the tomato that you eat is the fruit of the plant, sort of like you were the fruit of your mother's womb. Botanically it IS a fruit. But remember, so is a cucumber, green bean pod, pumpkin and zucchini. Most of us probably wouldn't make a fuss over whether a green bean or cucumber is a fruit or vegetable. Common sense says they are a vegetable.

Linguists, especially of the computational variety, have a lot to say about defining terms for common-sense reasoning; that's most of the point of building ontologies, which are essentially taxonomies of concepts. At each level of the ontological hierarchy, the idea is to partition concepts into non-overlapping categories. In principle, this ought to be easy to do. So, for example, we might do as SUMO does, and start by saying that everything that exists is an entity of some sort. All entities exist as either physical or abstract objects. All physical entities are either objects or processes. Once you've got the hang of this sort of partitioning, it's easy to swing away with Occam's axe, hacking the entire world into neatly-defined categories.

However, as we've written before, once you get past the very most abstract levels, it can be anything but obvious what the correct partions are. And even at those abstract levels, there's often substantial disagreement about terminology. This gets us to the root of the problem with tomatoes; when people argue about whether or not they're vegetables, they're really arguing about their choice of partitioning rules. We've got:

• The botanist rule: mature plant ovaries are fruits, everything else is a vegetable.
• The horticulturist rule: perennials are fruits, annuals are vegetables.
• SC's rule: if your mom had to force you to eat it, it's a vegetable

The U.S. Supreme Court took up the question in 1893, and decided that:

Botanically speaking, tomatoes are the fruit of a vine, just as are cucumbers, squashes, beans, and peas. But in the common language of the people, whether sellers or consumers of provisions, all these are vegetables...

In other words, they came down for SC's rule and ordinary language as the choice of partitions for plant products. The developers of the Cyc ontology, the world's largest collection of concepts, formalize it similarly:

(DEFINE-OKBC-FRAME vegetable-food :SUBCLASS-OF (food vegetable-matter) :INSTANCE-OF (class existing-stuff-type default-disjoint-food-type) :OWN-SLOT-SPECS ((documentation "A collection of edible stuff. Each element of Vegetable-Food is a foodstuff which is derived from a plant and is ordinarily considered a vegetable; e.g., a carrot (an instance of Carrot-Foodstuff), a potato (an instance of Potato-Foodstuff), a lima bean (a Bean-Foodstuff), a tomato (a Tomato-Foodstuff). Note: Vegetable-Food includes certain plant parts that are technically classified as fruits by botanists, but which are treated as vegetables in food classification -- such as tomatoes. These would, e.g., be found in the vegetables section of a supermarket, and they satisfy more of the axioms about vegetables than those about fruits (e.g., sweetness.)")) ) [emphasis added by me -- SC]

It can't be put more succulently succinctly -- tomatoes really do "satisfy more of the axioms about vegetables". Botanists may cringe, the editors of Reason may scoff, but from the standpoint of a linguist, three cheers for the New Jersey Assembly for standing up for common sense.

Comments

[kurlow]

loving the narrative, hope you are having fun!!!

--------------
http://www.masbromas.com/