Pancakes were so-called for a reason. They are cakes and they are cooked in pans. It doesn’t follow that they are so-called for a reason. Or, rather, if it follows, it does so only because they are called “pancakes” for the reason that they were called “pancakes” and they were called “pancakes” for a reason. So, even though they are now cakes that are cooked in pans, it doesn’t follow that they are now called “pancakes” for that reason.
Compare here the following. Dartmouth was so-called because it was at the mouth of the river Dart. But if the river Dart had changed its course over time so that Dartmouth was no longer at its mouth, the town could nonetheless have retained its name. In that case, it would be true that Dartmouth was so-called for the reason that it was at the mouth of the Dart, but it would be false that it is so-called because it is at the mouth of the Dart. So, even if it is now true that Dartmouth is at the mouth of the Dart, it doesn’t follow that it is so-called for that reason. Similarly, it doesn’t follow from the fact that pancakes were once so called because they were cakes cooked in pans, and the fact that pancakes are cooked in pans, that they are so-called because they are cooked in pans.
What follows from all that? One thing that follows is that one can’t now work out what pancakes are from the fact that they are called “pancakes”. A natural question to ask, then, is whether one could once tell what they were on the basis of what they were called. Specifically, could one tell what they were from what they were called when they were called “pancakes” because they were then cakes cooked in pans? Plausibly, one couldn’t. For all one could tell then, “pancakes” derived not from a combination of an expression for pans and an expression for cakes, but rather in some other way. For example, for all one could tell then, “pancakes” may have derived from the name of their inventor, Pancake. Pancake was so-named by her parents because they liked the way the name sounded; it was but a happy coincidence that she invented a cake made in pans.
Perhaps, however, one could then have been in a position to know what pancakes were on the basis of (i) knowing that they were called “pancakes,” and (ii) knowing that their name was a combination of an expression for pans and an expression for cakes. On that basis, one could at least figure out that they had something to do with pans and cakes. Would that be enough? Plausibly, it would not be enough. For all one could tell just on that basis, the things called “pancakes” might have been any of the following:
Cakes stored in pans.
Cakes stored with pans.
Cakes shaped to fit into pans.
Cakes shaped like pans.
Cakes used to clean pans.
Cakes made to celebrate pans. (Suppose, for example, that people celebrated the anniversaries of their purchases of especially fine pans.)
Cakes made of pans. (Suppose, for example, that some pans were made of a material that was edible by us, or that there were creatures that ate metal pans.)
Cakes made for wild pans to eat. (Suppose, for example, that some pans were alive.)
So, it’s plausible that no one has ever been in a position to tell what pancakes are just from their knowledge of what pancakes are called, even if their knowledge also encompassed the fact that “pancake” is a combination of an expression for pans and an expression for cakes. Nonetheless, pancakes were once so-called because they were cakes cooked in pans. It seems to follow that knowing that fact about pancakes, or “pancakes,” is different from knowing what pancakes are called, that their name is a combination of expressions for pans and cakes, and that pancakes are cakes cooked in pans.