Speak, memory?  

In an all too rare break from his new legal heaven/haven, Will Baude writes about senses and their relationship with memory.

[M]y own submission is that touch is an underrated and visceral sense, but mostly I find the insistence that there is one single sense most closely tied to all people's memories sort of bafflingly confident.

I agree that touch is underrated as a sense. But why couldn't there be some biological reason one sense is favored over another in memory? It turns out a friend of mine is a neuroscientist (at Yale, Will) who researches this very question, and her take seems to be that different senses arose at different evolutionary times, and are therefore connected more closely with functions that may have emerged concurrently. Touch was probably the first sense, followed by taste, smell, hearing, and finally vision; these are associated with successively higher order functions in the brain. So, looking at it from the other side, the fact that smell is so often strongly connected with memory and emotion may speak to questions of when memory and emotion arose, how they work, etc. Or, it might provide a basis for thinking of different kinds of memory, from an evolutionary neurological standpoint.

I don't think this has any particular bearing on the anecdotal and literary evidence Will cites, and I certainly don't think it precludes other senses from carrying memories. My own experience is the music brings with it by far the most present memories, and I'm moved by music more often (by at least an order of magnitude) than any other artform. But I know this isn't the case for everyone.