It is with a somewhat heavy heart that I commit these thoughts and feelings to paper. I have always been a fan of Slater’s approach to fieldwork
While Onesimus was busy blithering on about “linguistic purity” and related nonsense, I was busy reading Slater’s most recent and characteristically bold analysis, this time of Notso Yi (Slater 2012). I felt a twinge of recognition when I skimmed the data, and a twinge of embarrassment for Slater, and then a teeny tiny twinge of respect for Onesimus’s point, when I read these words:
No doubt similar processes will be discovered in other languages, now that I have alerted fieldworkers to the possibility of such systems.
No doubt, if Slater had done maybe just a little “deskwork”, just a quick literature search, he would have found my own work on the Winodanugai language, which offers a surprisingly similar type of grammatical class to Slater’s Notso Yi Enumerated Evidentials, namely Eidetic Pronouns. I’m not going to recapitulate all of the details of either Enumerated Evidentials nor Eidetic Pronouns
Long story short, both Enumerated Evidentials and Eidetic Pronouns require what, to most people, look like prodigious feats of memory. In the case of Winodanugaian Eidetic Pronouns, this is explained by their curious and curiously specific criminal ancestry (Searsplainpockets 1993). For Notso Yi Enumerated Evidentials, the mystery is as yet unexplained. (A side note to Onesimus and his ilk: one of the added benefits of fieldwork, at least to the linguistic anthropologist or anthropologically inclined linguist, is that you can take blood samples and do genetic testing! Welcome to the 21st century.)
While the Slater Method has much to recommend it, it is not for me, and one of its few but vital shortcomings demonstrates why; because the interaction between fieldworker and fieldworkee is mediated in English, the fieldworker develops no sense for the sound of the language. I would like very much to know about the phonology of Notso Yi, but I expect even Slater is not familiar with it.
In order to resolve these deficits, I will soon be employing the Malinowski Method of Fieldwork; that is, I will visit the Notso Yi, gather just enough of my own data, make my own bold analyses, and make my own far-
In the meantime, I need to point out that Slater’s analysis did make me reconsider my own analysis of Winodanugaian Eidetic Pronouns as potential evidentials, rather than as pronouns. I was able to make the data fit such an analysis, though it required several hyperparameters in the form of two distinct null morphemes and a silent phoneme with six silent allophones. Using an analysis of the Kolmogorov complexity of the competing analyses (completed by my good friend and new computational colleague Chit Fullah (see Fullah 2010)), I was able to determine, by comparing the alternative Minimum Description Lengths, that my original analysis was 32.77% superior to the new one. (Good try, though.)
• Fullah, Chit, 2010, “How Computers Can Do Fieldwork For You: A Case Study,” SpecGram, CLVIII.4.
• Onesimus, H.D., 2012, “The “Slater Method” of Linguistic Fieldwork,” SpecGram, CLXIV.1.
• Plaid’oh, Fædrus Ϙ. Χ., 2009, “On Google, Wikipedia, and The Development of The Internet,” Collateral Descendant of Lingua Pranca.
• Searsplainpockets, Claude, 1993, “Eidetic Pronouns: An Anthropological Linguistic Study of the Winodanugai,” SpecGram, CXLVII.3.
• Slater, Keith W., 2012, “Directional Source-