I was briefly excited by the title of a recent Language Log post, Earworms and White Bears, thinking it might have something to say about, well, worms that people put in their ears. However, we immediately learn that the earworms in question are simply catchy tunes that get caught in people’s minds.
My excitement, though, stems from the fact that the Nantis of southeastern Peruvian Amazonia, with whom I have worked a bit (see here and here), actually do sometimes put worms — or more precisely, larvae — in their ears. I’ve never heard or read about any other group that makes use of larvae in this way, though, so I was momentarily hoping that Language Log would change that
The Nantis call the larvae in question magempiri, and they are one of a number of larvae that Nantis help flourish by means of a form of low-intensity animal husbandry where they puncture the trunks of the palm species in which the relevant species of beetle must lay their eggs. When the magempiri are the right size, a portion of the trunk is split open and some larvae removed, together with some of the pulp on which the larvae are feeding. The pulp and larvae are then wrapped up in Heliconia leaves, in which the larvae can live for several days.
The magempiri are used to clean one’s ears: tilting one’s head, one drops a larva into the ear canal and the magempiri then starts munching away on what it finds there, creating an incredible racket (for the temporary host), and producing a funny if somewhat gratifying ticklish feeling. When the magempiri gets full, it becomes inactive, and simply falls out of the ear canal when the user tilts his or her head the opposite way. Repeat until satisfied. All in all, this is one of the most fascinating uses of an insect species that I have come across in the Amazon Basin.
And there is even an interesting story to be told about the name of the larva. The word is a derived from the root magempi ‘be deaf’ with the agentive nominalizer -ri, so that the name might be rendered literally as ‘deafener’ — presumably a reference to the noise the user hears while the larva is doing its work. Interestingly, it turns out that this verb root is important because it is one of the few places in Nanti where a remnant of the Proto-Arawak privative ma- survives (for a brief discussion, see here). In many Arawak languages, ma- still functions as a privative, deriving a denominal stative verb (for a comparative discussion, see here). In Nanti, however, the only evidence that the privative was once productive in the relevant branch of the Arawak family are these frozen traces of the privative. For the Nanti word in question, the gempi of ma-gempi is obviously related to modern gempita ‘ear’, so that magempi would presumably have meant ‘be lacking ears’.
I’d be really interested to know if any other Amazonianists have encountered real earworms, and not the metaphorical ones discussed on Language Log (discussion continues here).