One of my chief interests is comparative Chibchan linguistics, and the historical relationships between the attested languages. Besides the masterful and fundamental Constenla Umaña Comparative Chibchan Phonology dissertation of 1981, another key (albeit smaller) work for Magdalenic Chibchan and the Arhuacan languages in particular is Frank's "Proto-Arhuacan Phonology" of 1993. Frank's work represents some initial efforts at reconstructing morphemes from the common ancestor of the modern Arhuacan languages and laying out some of the sound changes for the different languages. Of course, Frank's efforts were complicated by the then-even-poorer state of knowledge about Arhuacan vocabulary, and he was working principally by gleaning vocabulary from earlier discussions of the languages and their phonology. Inevitably, there are some uncertainties and inconsistencies in Frank's own analysis (some updating is provided by Jackson 1995), but it remains an important starting point for further work and its list of cognates and reconstructed forms has hardly been bettered.
More recently, additional publications concerning the Arhuacan languages have more information about their vocabulary and phonology (though we are talking about a field in which the publication of a few words still has a significant percentage impact). Of special note are Landaburu 2000 and Ortiz Ricuarte 2000 (both in the mighty González de Perez & Rodríguez de Montes 2000 collection), and Trillos Amaya 1999. After 15 years, it may be worth beginning to re-examine the Arhuacan material and see if there are new insights to be gained regarding the relationships between and histories of these languages.
With this in mind, I have been considering the various words for "snow" and "ice" in the various Arhuacan languages.
Beginning with Frank 1993, he compares three forms for "snow" -- Ika dʒʌn, Kogi nu'wabi, Damana dɨm -- and proposes a Proto-Arhuacan *dub. Frank's cited modern forms are largely the same as those reported for "snow" in Huber & Reed 1992, a work which additionally offers forms for "ice": Ika dʒwábu, Kogi nəbbu-gəldə, Damana dɨmɨ-ngɨna. The Kogi form here is broadly confirmed by the Kogi form /nab'gala/ (/l/ being realized as [lʒ] intervocalically, explaining the Huber & Reed spelling) in Ortiz Ricuarte 2000, which additionally gives the Kogi "snow" word as /nu'abi/ which, though Ortiz Ricuarte states /u/ is only realized as [w] in syllable initial position, seems effectively the same as Frank's nu'wabi and (and Huber & Reed's nuwábi). In a partial Swadesh list for Ika, Landaburu gives abitsi for "snow", but this seems to be perhaps related to an Arhuacan word for "white" (Ika bunsi, Kogi /a'bunci/, Damana ambɨnʃi). However, in his discussion of Ika phonology, Landaburu also provides two additional different words for "snow": dʒewə (p. 734) and dʒəN (p. 736, where "N" is a nasal of variable realization, in this case probably [ŋ]); the latter form seems comparable to Frank's dʒʌn (Huber & Reed's dʒən), though it is difficult to account for the apparent contrasts between /n/ or /N/ and /w/ in the former version without more information. At present, I have not isolated any descriptions of Ika words for "ice", nor of the Damana words for "snow" or "ice" from any of Trillos Amaya 1998, Trillos Amaya 1999, or Trillos Amaya 2000.
Paucity and inconsistency of data remain serious challenges to those researching Chibchan languages.
However, returning to considering the relationships between the various Arhuacan words for "ice" and "snow", it seems to me like we now now enough to consider revising Frank's original reconstruction of Proto-Arhuacan *dub for "snow".
To begin with, it seems likely that Kogi nabˈgala “ice” is directly cognate with Damana dɨmɨ-n-gɨna. Both seem to share a common suffix, Kogi -gala and Damana -gɨna, as Frank identifies intervocalic Kogi /l/ and Damana /n/ as common reflections of a Proto-Arhuacan intervocalic */d/ (the -n- preceding the Damana suffix is a connective element, actually a realization of an archiphoneme /N/; Trillos Amaya 1998). The complete proto-form of the suffix may have been something like *-gədə. Ortiz Ricuarte 2000 describes the Kogi -gala suffix as providing the sense of "matería con que está hecho", offering the examples of /ʃei'ʒa/ "machete" and /ʃei'ʒagala/ "iron". Accepting this analysis (while remembering that there are less clear examples of the same suffix and that iron machetes are Spanish-era additions to Kogi culture), nab'gala "ice" would be "the stuff from which from nab- is made", suggesting strongly that we could understand Kogi nab- here as "snow", cognate with Damana dɨm, and Ika dʒeN. Frank gives us further confidence regarding this equation, as he shows that Proto-Arhuacan initial */d/ produced Kogi /n/ and, before high vowels, Damana /d/ and Ika /dʒ/, while Proto-Arhuacan final */b/ gave Kogi /b/, Damana /m/, and Ika /n/.
Thus, it seems questionable as to whether Kogi nu'abi really is a direct cognate of Ika dʒəN or Damana dɨm as Frank seemed to be assuming. Moreover, the vocalism of Kogi nab- seems unlikely to procede from *dub. Instead, a proto-form *dəb- would better account for Ika dʒəN, Kogi nab-, and Damana dɨm.
However, it should be noted that a Proto-Arhuacan *dəb- "snow" would be unexpectedly similar to the Constenla Umaña's reconstructed Proto-Chibcha *dəbə, "feline" that likely produced Kogi ˈnabi “feline, jaguar” as well as Muisca nymy “gato montés”; Diccionario y gramática chibcha 1987). Other apparent possible cognates of Proto-Chibchan *dəbə may be found in Damana dumá-ga “jaguar” and perhaps the second element in Ik seiku-númɨ “jaguar” (both cited in Huber & Reed), though here in the "u" vowels are is difficult to explain, as I can think of no parallels for PrCh */ə/ > /u/ unless the Ika and Damana "u" in these forms is meant to represent /ʉ/ as Jackson writes in Ika "kʉnʉ", Damana "kʉna" (both “leg” ) for which Frank gives respectively /ʌ/ or /ɨ/ in Ika "kʉni", Damana "kɨna" (both “leg”). Unfortunately, I can find no examples of words for "jaguar" in the more recent reports regarding Ika and Damana of Landaburu or Trillos Amaya, so the issue of the vocalism in these words remains slightly mysterious to me.
Ortiz Ricuarte (1998 & 2000) has suggested that Kogi nu'abi might possibly have been borrowed from Spanish nieve, but this seems unlikely to me. Inhabitants of the Sierra Nevada de Santa Marta are likely to have had a word for snow before the Spanish arrival and, though it is difficult to suggest likely datings for reconstructible Proto-Arhuacan forms, in all likelihood they belong to a pre-Spanish epoch. Moreover, it is difficult to account for the change in vocalism of Spanish /ie/ to Kogi /a/.
Therefore, it seems clear to me that we must regard the Kogi nab'gala and Damana dɨmɨngɨna (both "ice") as suffixed forms of a proto-word for "snow" that survives directly in Damana dɨm and Ika dʒəN (whether from a form like *dəb- or with some other vowel). It also seems likely that the attested Kogi word for snow, nu'abi, is directly cognate with reported Ika dʒwábu “ice”, perhaps from a proto-form like *duab-, evidently distinct from (but not necessarily unrelated to?) Ika dʒəN, Da dɨm, and presumably, the Kogi element nab- in nab'gala. Unfortunately, as yet I have found no additional examples in the modern Arhuacan languages that might illuminate the developments of Kogi nu'abi and Ika dʒwábu. Clearly, more detailed information about the phonologies of the various modern Arhuacan words involved would help in unraveling the history of the forms more completely.
Works Cited
Constenla Umaña, Adolfo, Comparative Chibchan Phonology, Ph.D. dissertation (Philadelphia: Department of Linguistics, University of Pennsylvania, 1981).
Diccionario y gramática chibcha: manuscrito anónimo de la Biblioteca Nacional de Colombia, ed. by María Stella González de Pérez (Bogotá: Instituto Caro y Cuervo, 1987).
Frank, Paul S., "Proto-Arhuacan Phonology", Lingüística Chibcha 12 (1993), pp. 95-117.
González de Perez, María & María Luisa Rodríguez de Montes, eds., Lenguas indígenas de Colombia: una visión descriptiva (Bogotá: Instituto Caro y Cuervo, 2000).
Huber, R.Q & R.B. Reed, Vocabulario Comparativo(Bogotá: Instituto Lingüístico de Verano, 1992).
Jackson, Robert T., "Fonología comparativa de los idiomas chibchas de la Sierra Nevada de Santa Marta", Boletín Museo del Oro 38-39 (1995), pp. 57-69.
Keller, Natalia Eraso, ed., El léxico del cuerpo humano a través de la grámatica y la semántica, Lenguas aborígenes de Colombia: Memorias 5 (Bogotá: Centro Colombiano de Estudios de Lenguas Aborígenes, 1998).
Landaburu, Jon, "La lengua ika" in Lenguas indígenas de Colombia: una visión descriptiva, ed. by María Stella González de Perez & María Luisa Rodríguez de Montes (Bogotá: Instituto Caro y Cuervo, 2000), pp. 733-748.
Ortiz Ricuarte, Carolina, "El léxico del cuerpo humano en la lengua kogui: Hacia una clasificación de los formantes nominales" in El léxico del cuerpo humano a través de la grámatica y la semántica, ed. by Natalia Eraso Keller, Lenguas aborígenes de Colombia: Memorias 5 (Bogotá: Centro Colombiano de Estudios de Lenguas Aborígenes, 1998), pp. 15-34.
Ortiz Ricuarte, Carolina, "La lengua kogi: fonología y morfosyntaxis nominal" in Lenguas indígenas de Colombia: una visión descriptiva, ed. by María Stella González de Perez & María Luisa Rodríguez de Montes (Bogotá: Instituto Caro y Cuervo, 2000), pp. 757-780.
Trillos Amaya, María, "Léxico del cuerpo orgánico en damana" in El léxico del cuerpo humano a través de la grámatica y la semántica, ed. by Natalia Eraso Keller, Lenguas aborígenes de Colombia: Memorias 5 (Bogotá: Centro Colombiano de Estudios de Lenguas Aborígenes, 1998), pp. 35-57.
Trillos Amaya, María, Damana, Languages of the World/Materials 207 (München: LINCOM Europa, 1999).
Trillos Amaya, María, "Síntesis descriptiva de los sistemas fonológico y morfosintáctico del damana" in Lenguas indígenas de Colombia: una visión descriptiva, ed. by María Stella González de Perez & María Luisa Rodríguez de Montes (Bogotá: Instituto Caro y Cuervo, 2000), pp. 749-756.
