You can't have it both ways. The only affect that the market has on the local market for lobsters in Bangladesh, is via the price in Bangladesh, i.e. the price the fishermen pay.
So to the extent that they are affecting the market in Bangladesh, they must be paying higher prices, and therefore the parent comment applies.
>But unless you're an economist, you have to posit that the enjoyment from a lobster dinner is the same whether you're a Bengali villager or a yuppie in San Francisco. In a sense they're giving up $60 of enjoyment to get a couple of dollars of cash.
The economists are in fact correct. You cannot measure enjoyment in dollars, because $60 will buy a lot more additional enjoyment for a Bengali villager than a San Francisco yuppie. So the villager would be willing to give up the enjoyment of a lobster dinner for much less than $60, which is precisely what they are doing.
You could also make the argument that the villagers have the opportunity to have $60 of enjoyment at a bargain price, whereas the San Francisco yuppies must pay a high price.
Huh? I could take a friend to a restaurant and order two of the same dish. Odds are sky-high that enjoyment from eating the same dish, prepared in the same restaurant, against the same cultural background, would be rather different.
In neoclassical economics, we assume rational actors. So it must be 2. But why would a Bengali villager get much less utility out of a lobster dinner than a San Francisco yuppie? Sesfood is a staple in Bangladesh, and lobster is a delicacy. [1] The answer is because they're poor: measured by the market, they, ostensibly, ascribe less value to everything than the San Francisco yuppie.
That's an interesting result. The San Francisco yuppie derives far more utility from that lobster than the Bengali villager does. Why? Because he's richer. By the measure of the market, the rich derive far more utility from everything, and the market allocates scarce resources to those who value them the most.
[1] This is not an example out of an economics textbook, where country A sells something it doesn't value to country B who does value it.
Edit/Clarification: the essence of social welfare economics is in trying to do interpersonal utility comparisons that yield a social utility function that behaves as much as possible like a personal utility function. Arrow's theorem tells us that this is not possible to do perfectly; any attempt to do so will result in one of several significant deviations from what we consider "utility" for a single person.
Utility is a very specific tool devised to model choice behavior. It is a basic idea of utility theory that "Vikram gets much less utility from the lobster than Yolanda" is not a meaningful statement within the framework. There's no wrangling or wiggle-room with this point; interpersonal utility is formally incomparable.
Similarly, it is well-trodden that markets tend to allocate scarce goods to those with the greatest willingness and ability to exchange other scarce goods for them; market theory doesn't presume to say anything about who enjoys the scarce goods the most. This is why, beyond choice and market theory, we need something that addresses social value systems. Utility does not pretend to do this, except to the extent that an individual may value social justice in their personal utility function.
So, while I take your point that my post was technical and, yes, dry, them's the ropes when it comes to utility theory. Correct statements tend to be boring ones that don't evoke human interest. If one wants to evoke human interest (a perfectly reasonable goal) I think it's best not to mix in half-appropriate technical concepts when there is substantial potential for confusion.
[0] For anyone who is curious, I estimate 80% of Ariely's "irrational" examples are easily explained by understanding that making optimal choices with imperfect information has a cost in time and effort; when you price in this cost in time and effort, making suboptimal-but-easy-to-make or even random choices is actually perfectly rational within the framework.