chenille wrote:It doesn't actually have anything to do with the straights. It has to do with the relatively narrow isthmuses
I'm sorry, I caused unnecessary confusion here by saying straight when I meant isthmus.
There are no straits in Panama or Suez. There are canals, now, which aren't the same thing anyway and don't count because they're just things we built across the land and not a substantial alteration of it. I meant a sufficiently narrow isthmus counts as a discontinuity in an otherwise continuous mass of land (meaning dry land, not covered by water).
Also, going by Max's tectonic plates definition having nothing arbitrary, what stops either southern New Zealand or western California (whichever has a larger portion on the Pacific plate) from counting as The Pacific (Sub-)Continent? For that matter, why don't they both count as a Pacific Subcontinent of the Australian and North American continents, respectively? Or if we're just going by largest land mass completely contained within a tectonic plate... well in that case Eurasia and North America are out since they both cross over onto other plates, but why isn't the Big Island of Hawaii the Pacific Continent? Point being if you're not counting every island (or at least the largest land mass on each plate regardless of size) as a continent, and every part of a land mass on a different plate as a different sub-continent, then you are drawing some arbitrary line. I'm just drawing one other arbitrary line. A continent is a continuous land mass: we agree that the mass (rather surface area) must be of sufficient size, and I'm adding that the continuousness also requires a certain size. Unless you can propose some non-arbitrary way of picking those size cutoffs, we may as well pick arbitrary ones that match common usage. Or pick none and call every island a separate continent (and coastal California its own subcontinent, etc).
I'm also looking strictly at physical geography here; I don't buy anything about Europe or Central America being different "continents" because of cultural boundaries. But the physical attributes in question when discussing continents are continuous masses of "land" in the sense of crust not covered by water, as compared to crust covered by water; and tectonic plates fundamentally are not about that. Tectonic plates surely are another notable type of object in physical geography, but are fundamentally different from continents and only bear a tenuous and by no means total coincidence them (meaning: yes the bulk of a given continent tends to reside on one plate and there are known causes for that coincidence, but as all these exceptions show there is not always or even frequently a one-to-one correspondence between continents and plates).