Volume 95 1986 > Volume 95, No. 2 > Micronesian dialects and the overnight voyage, by J. C. Marck, p 253-258
                                                                                             Previous | Next   

- 253
JEFFREY C. MARCK Drake University

One day I was preparing a visual aid for a class I was teaching on Micronesian languages. Since there is so much water and so little land in Micronesia, I decided to draw radii out from the inhabited islands to give the students a notion of the distribution of land, people and languages through the area. I first drew 200-mile radii and produced what would have been a fairly useless classroom aid. It showed only that most islands are within 200 miles of at least one other and each area blended into the next without much visual effect. I next tried 100-mile radii and as I worked from east to west across the map, the language and dialect groups emerged in a clear pattern as I progressed (Figure 1).

Ethnographic literature on the sailing technology (Gladwin 1970, Lewis 1972) has taught us that: (1) 100 miles is the approximate limit of the overnight voyage, i.e., if one leaves a certain island at dusk one day, landfall can reliably be made in daylight of the next day only if the target island is less than 100 miles away and the winds are favourable in direction and strength; (2) senior navigators are capable of any number of very lengthy voyages but junior navigators rarely go out alone on voyages of any great distance; and (3) the overnight voyage is often spontaneously initiated by intermediate and even junior navigators who have no purpose other than to visit family, friends or sweethearts, and they regularly, safely transport full loads of passengers when they do so.

So, it seemed to me that the sailing technology created a certain opportunity within the social system that manifested itself linguistically in approximately the following manner: (1) people within a day's (24-hour) voyage of another island maintained patterns of social interaction with that island's inhabitants that resulted in maintenance of mutual linguistic intelligibility between the two populations; and (2) the role of average or nonsenior navigators (the spontaneity and regularity of their journeys over the millennia) resulted in different kinds of social, economic and linguistic continuity between those islands separated by more than a single night at sea and those separated by less than that distance.

Local Manifestations

The local manifestations of this pattern are as follows. This exposition is based upon exposure to scholars working on these languages, my own work on the languages (e.g., Marck 1975 and 1977) and many readings of Bender (1971), which is the main publication to date with materials on the language boundaries

- 254
- 255

and dialects through the area. Bender (1984), the most recent set of works on Micronesia, is most concerned with problems in the history and description of the various languages. A few of those articles are cited below as they relate in one manner or another to the language diversity question.

  • 1. Kiribati (Gilbertese). A single language with only very moderate north/south pronunciation and lexical differences. The lexical differences seem mainly a result of borrowings from Tuvalu (Ellice) in the south.
  • 2. Nauru and Banaba. Nauruan is a very distinct language. Banaba has a recent history of intensive social and political interaction with the Gilberts and I am not certain what the linguistic interaction has been, how deeply into prehistory it may have extended and in what interactional matrix. In terms of the nature of the Banaba language, I have simply heard that it is “different” (G. Groves, personal communication).
  • 3. The Marshalls. Eastern and western dialects in the respective island chains with minor, but possibly ancient, differences in pronunciation and the lexical inventory. Ujelang and Eniwetok dialects (and social structure) are said to be noticeably different from those of Bikini or Kwajelein (R. Kiste, personal communication).
  • 4. Kosrae. A single language with no nearby dialects that may have had some dialectal variability before a depopulation episode after European contact (Lee and Wang 1984).
  • 5. Ponape, Ngatik, Mokil and Pingelap. These islands have dialects that hover around the lexicostatistical measure of mutual intelligibility with all immediately adjacent islands. (See Rehg 1984 for comments on dialects on the island of Ponape.)
  • 6. Nukuoro and Kapingamarangi. Very distinct Polynesian languagues which are not mutually intelligible.
  • 7. Truk and Trukic. Bender (1971), Jackson (1983, 1984) and Quackenbush (1968) all contain material relevant to this, the most complex dialect and language situation in Micronesia. By the overnight voyage criteria we have six theoretical areas of interactional intensity: i. the Truk area; ii. the Woleai area; iii. the Ulithi area (Ngulu would be outside the area were it not for Yap as an intermediate stop, for although Yap is not Trukic, it allows for shorter, safer voyages and more sponaneity in exchange of visits between Ngulu and the other areas—see Map, Figure 1); iv. the Sonsorol area; v. Merir; and vi. the Tobi area.
  • Mapia once had a Trukic-speaking population but it was decimated by European slaving raids in the 19th century and little is known of the language except, through short word lists, that it was Trukic (Bender 1975, Jackson 1983).
  • The Truk area is a chain of dialects that are truly mutually intelligible with those of immediately adjacent islands. It is of interest to the present discussion that it is only those islands separated by more than an overnight voyage that do not have mutually intelligible languages. Comparative linguistic criteria indicate the Truk area to be the probable homeland of all Trukic and oral traditions state that this was the case.
- 256
  • The Woleai area is less diverse, internally, than the Truk area and aggressive local traditions of navigation appear to have resulted in a situation where a high degree of mutual intelligibility is found between Satawal and the Pulusuk/Puluwat area.
  • The Ulithi area dialects are very similar internally and close to those of Sonosorol, Merir and Tobi which may have differentiated so recently that natural divergence processes have not proceeded to any great extent.
  • 8. Yap. A linguistic isolate with a different (pre)historical progenitor than surrounding languages (i.e., in its last several thousand years of history, though ultimately, all Micronesian languages are descended from Proto Austronesian).
  • 9. Belau. A linguistic isolate with a different (pre)historical progenitor than surrounding languages.
  • 10. Marianas Chamorro. Rather nothing is known of dialects north of Rota before the populations’ decimation by Europeans at the end of the 17th century. Rota has a dialect very much like those on modern Guam and archaeological evidence through the area suggests the kind of broad continuity typical of an archipelago participating in a common culture although mutual intelligibility of local languages would not have been necessary to sustain such a pattern (e.g., Tonga and Samoa).
A Simple Rule

A simple rule can be stated in partial summary of these observations:

If two islands are separated by a voyaging distance involving a single night at sea or less in traditional craft, their dialects will be mutually intelligible. The maximum distance is about 100 miles in Micronesia.

The rule must be modified to dismiss such cases as the comparison of Yapese to the Ngulu Trukic dialect, since they derive from different prehistoric arrivals of people and languages into the area. That is, the rule relates to limits towards which dialects of a language are expected to diverge, not the extent to which languages not immediately related are expected to converge. Other than that, the rule appears to need no other modification and there are no exceptions.

There are cases in the western Trukic dialect area where dialects may be mutually intelligible across greater, even rather vast distances, e.g., Ulithi and Sonsorol, but that case most certainly is the result of a fairly recent physical dispersal of people rather than a special, local ability to maintain a low level of language differentiation over long distances. Also, eastern and western Marshallese dialects show very little differentiation and within either of the dialects, local differences are normally said or inferred to be absent (see Bender 1969, 1971). The Marshallese situation does not violate the rule, it simply presents a case where linguistic homogeneity is found within a larger area than anywhere else in Micronesia. The rule states that diversification will not proceed beyond a certain point (i.e., a breakdown of mutual intelligibility). It does not state or attempt to explain what conditions prevent diversification.

- 257

Although I noticed the likely socio-technical basis for the rule by accident, the regularities to be observed are consistent with the notions outlined by Pawley and Green (1973), reinforced by such work as Lensink (1976) and continued more recently by Pawley (1981) suggesting that geography, technology and length of time since initial settlement have created regular patterns of linguistic continuity and differentiation throughout the island world where Oceanic Austronesian is spoken. The present rule can be stated for Micronesia without reference to time since initial settlement.

The archaeological data base is too marginal in Micronesia to completely compare the pattern to Pawley and Green's model, Lensink's observations about eastern Melanesia or Pawley's more recent elaboration of the Pawley-Green model. But the linguistic and ethnographic data (e.g., Bender 1971, 1984; Bender and Wang 1982; Bender et al. 1982; Jackson 1983) are now well enough developed to contribute substantially to an expansion of Marck's (1975) settlement model, and to help frame culture change questions in a manner that would enhance the problem orientation and cost effectiveness of archaeological strategies for the area. The roles of time and demographic interactions are especially important and I hope that the rule stated above will add some small insight into the variables that interacted in the process of the Austronesian dispersal and subsequent cultural differentiation.

By the addition of the overnight voyage rule and other such observations to our inventory of totally consistent phenomena in Oceanic situations, perhaps we can support and expand Pawley and Green (1973) and Pawley's (1981) hypotheses that numerous geographic phenomena determine the degree of linguistic unity or differentiation (see Lensink 1976), and that this, in turn, creates certain social parameters within which ethnic and political unity emerges or deteriorates (see Pawley 1981).


An early version of this paper was presented at a symposium on Micronesian languages at a meeting of the Linguistic Society of Hawaii in 1978. The author wishes to thank Mac Marshall and Andrew Pawley who made many useful comments on a recent draft of this paper.

  • Bender, Byron W., 1969. Spoken Marshallese. Honolulu, University of Hawaii Press. PALI Series.
  • —— 1971. Micronesian Languages, in T. E. Sebeok (ed.), Current Trends in Linguistics, Vol.8, Oceania, pp.426-65. The Hague, Mouton.
  • —— 1975. Mapian: An Extinct Micronesian Language? Paper presented at the Second Annual Oceanic Linguistics Festival, Linguistic Society of Hawaii, Honolulu.
  • —— (ed.), 1984. Studies in Micronesian Languages. Canberra, Pacific Linguistics, Series C, No.80.
- 258
  • —— and Judith W. Wang, 1982. The Status of Proto-Micronesian. Paper presented at the Austronesian Symposium, XVth Pacific Science Congress.
  • —— et al., 1982. Proto-Micronesian Word-list. Computer printout. Department of Linguistics, University of Hawaii.
  • Gladwin, Thomas, 1970. East is a Big Bird: Navigation and Logic on Puluwat Atoll. Cambridge, Harvard University Press.
  • Hollyman, Jim and Andrew Pawley (eds), 1981. Studies in Pacific Languages and Cultures in Honour of Bruce Biggs. Auckland, Linguistic Society of New Zealand.
  • Jackson, Frederick, 1983. The Internal and External Relationships of the Trukic Languages of Micronesia. Ph.D. dissertation. Linguistics, University of Hawaii.
  • —— 1984. Reflexes of Proto-Oceanic in the Trukic Languages of Micronesia, in Bender (ed.), pp.259-80.
  • Lee, Kee-dong and Judith W. Wang, 1984. Kosraean Reflexes of Proto-Oceanic Phonemes, in Bender (ed.), pp.403-42.
  • Lensink, Stephen C., 1976. Environmental Factors Affecting the Areal Size of Regional Populations in Melanesia. M.A. thesis, University of Iowa.
  • Lewis, David, 1972. We, The Navigators: The Ancient Art of Landfinding in the Pacific. Honolulu, The University Press of Hawaii.
  • Marck, Jeffrey C., 1975. A Lexicostatistical Model for Nuclear Micronesian Origins, Dispersal and Post-colonization Cultural Interaction. Proceedings of the Central States Anthropological Society, 1:1-22.
  • —— 1977. A Preliminary Manual of Nuclear Micronesian Sound Correspondences. Paper presented to the Austronesian Symposium, Linguistic Institute of the Linguistic Society of America, University of Hawaii, Summer, 1977.
  • Pawley, Andrew, 1981. Melanesian Diversity and Polynesian Homogeneity, a Unified Explanation for Language, in Hollyman and Pawley (eds). pp.269-309.
  • —— and Roger Green, 1973. Dating the Dispersal of the Oceanic Languages. Oceanic Linguistics, 12:1-67.
  • Quackenbush, Edward Miller, 1968. From Sonsorol to Truk: A Dialect Chain. Ph.D. dissertation. Department of Linguistics, University of Michigan.
  • Rehg, Kenneth, 1984. On the History of Ponapean Phonology, in Bender (ed.).