The Australian National University, College of Asia and the Pacific, CartoGIS

Interpretive essay: Linguistic issues and theories

The origin of Guam’s indigenous people has been a matter of considerable speculation for more than a century. Scholars have developed theories about CHamoru origins based on various evidence: physical (pottery shards, DNA, etc.), ocean movement and language.

Based on what is now known from linguistic studies coupled with the modern dating techniques of archaeology, scholars can say with some certainty that:

  • The original CHamorus spoke an Austronesian language that was an immediate descendant of Malayo-Polynesian or Extra-Formosan, the language that developed in the northern Philippines following the first migration out of Taiwan into the Philippines.
  • The earliest CHamorus sailed the 1,300 miles from the northern Philippines to the Marianas in the first major, open-sea migration in the history of mankind.
  • CHamoru is not close to any Philippine language, or to the Philippine group of languages as a whole, neither does it subgroup with any language group to the south of the Philippines.

Placing CHamoru proves difficult

Language is one of the key sources of information that enables scholars to draw conclusions about where people came from and to whom they may be related. This information, supplemented by the modern dating techniques of archaeology, can provide us with a fairly reliable picture of prehistory, although access to new data and dating techniques often require the prehistorian to revise theories that have become popular and are frequently cited.

CHamoru is one of the more than 1,200 members of the great Austronesian language family, a fact that has long been recognized. Austronesian is the family of languages that first developed in Taiwan some 5,000 years ago, and spread from there through the Philippines, to Indonesia and Malaysia, and ultimately into all the habitable islands of the Pacific, with some groups settling in mainland Southeast Asia and as far east as Madagascar.

The difficulty in determining the position of CHamoru relative to other Austronesian languages is at least partly the result of the complex history of contact with other languages, both Austronesian and non-Austronesian, which has taken place at an ever-increasing rate over the last 3,500 years or more since the islands were first settled.

These contacts have ranged from chance settlement by shipwrecked or drifting sailors from Micronesian islands to the south and east, or from any of a number of Philippine islands to the west or Indonesian islands to the south, to interaction with trade networks with other island groups, all of which have left their imprint on the language spoken by the indigenous population of the Marianas.

Impact of colonization

In the historical period, colonization under Spanish, German, Japanese and American governments along with decimation of the local population have had drastic effects upon the inherited language, especially in the lexicon, but also in the forms of words and the grammar. Probably thousands of words which were once part of the language of the early migrants have been lost, being replaced by words commonly used by the colonizers.

When attempting to determine the origins of the CHamoru-speaking people, it is a relatively easy task to identify and eliminate from comparison those terms which obviously have their source in Japanese or European languages. It is not so easy, however, to distinguish words that must have been inherited from the vocabulary of the first CHamorus from words that have been adopted into the language from other Austronesian languages.

For example for 150 years, starting in 1575, the Manila Galleon stopped annually in Guam for reprovisioning on its trips across the Pacific between Mexico and the Philippines, bringing with it not only commercial products for sale or barter, but Filipinos speaking a number of different languages, terms from which inevitably found their way into the vocabularies of the local CHamorus. The Spanish, in order to reinforce their own troops from Mexico, even recruited a company of Kapampangan-speaking soldiers from the Philippines, many of whom married local CHamoru women and introduced words from their language into CHamoru.

The techniques of historical-comparative linguistics, however, provide a principled method for separating out such borrowed terms from those that are inherited from the ancestral language. In order to understand these techniques it is necessary to briefly outline the principles upon which they are based.

All languages change

There are two basic assumptions upon which the principles and procedures of historical-comparative linguistics are based. The first of these is that all languages change over time. This seems a fairly obvious assumption when one compares the languages of grandparents with their grandchildren, but the cumulative results of such inter-generational change typically result in very different forms of words and grammar over periods of hundreds of years, as can easily be attested by comparing say the language of Chaucer or Shakespeare with the English spoken today. The expectation then is that CHamoru words that are directly inherited from the earliest ancestors who first set foot in the Marianas thousands of years ago are likely to be very different in form from the way they are spoken today.

At the point when ancient CHamorus first arrived, it can be assumed that they were speaking the same language as their relatives and friends whom they left behind. In the generations that followed, however, the changes that took place in the language of the homeland were independent of the changes that took place in the language of the CHamorus. This would have resulted in what would have been dialectal differences over a few hundred years, but ultimately, should the descendants of the travelers have returned to their homeland and attempted to communicate with the descendants of those who stayed behind, they would not have been mutually comprehensible. The two forms of speech would have become different languages, or “daughter” languages of their parent, or “proto-language”. Thus English and German are two of the daughter languages of their common parent language, Proto-Germanic.

Languages change systematically

The second basic assumption is that the sounds of a language change over time in regular and systematic ways, enabling linguists to establish sets of corresponding sounds between languages belonging to the same language family. Thus words in CHamoru with “f” correspond to similar words in Tagalog with “p”, while those with “p” correspond to Tagalog words with “b”. They can discover sets of words in these languages which must have had a common origin in the parent language because of their corresponding sounds and because they share the same or similar meanings. Such forms are known as “cognates.”

The discovery of such cognate sets enables the linguist to determine the sound system of the proto-language and to postulate the forms of the words in the parent language from which the daughter languages developed. This is not to say that idiosyncratic or irregular changes do not occur, they do, but frequently it is possible to determine regular processes by which the apparently irregular forms developed.

Words that are not inherited from the parent language, but have been adopted into the language by “borrowing” from some related language are often identifiable because they have not undergone the regular sound changes that characterize inherited words. Thus CHamoru babui ‘pig’ is clearly a borrowing of a word from one of the Philippine languages, many of which have babuy ‘pig’, because otherwise it would be pronounced in CHamoru as papui.

When languages share a set of sound changes or other innovations that have taken place, they can be grouped together into a subgroup, the members of which are more closely related to one another than to any language outside of that subgroup. So when comparing CHamoru with other Austronesian languages, the question that must be asked is whether there is a subgroup of Austronesian languages to which CHamoru is most closely related? If so, it should be possible to identify the homeland from which the original migrants set sail in their voyage of discovery to the Mariana Islands.

Some of the sound changes that characterize inherited CHamoru words provide evidence that CHamoru is in fact an Austronesian language.

CHamoru as an Austronesian language

There have been several studies which have focused on the systematic way sounds have changed in CHamoru which differentiate it from other Austronesian languages. The first of these studies was by the American philologist Charles Everett Conant, who spent many years in the Philippines at the beginning of the twentieth century. In articles published in the Journal of the American Oriental Society in 1907 and in 1911, and in Anthropos in 1911, he noted the relationship of CHamoru to Philippine and Indonesian languages, and attempted to describe how CHamoru consonants had changed over time.

The next major study which included a description of sound changes in CHamoru was by Hermann Costenoble, who spent his childhood years from 1905-1913 in Guam and grew up speaking the language. Writing in German under the tutelage of the great Austronesian scholar, Otto Dempwolff, he produced in 1940 a volume, Die Chamorro Sprache.

This work was ground-breaking for its time, but has since been superseded by more recent studies which have benefited from a fuller understanding of the nature of the parent language, Proto-Austronesian, and the far more extensive set of reconstructed forms with which comparisons can be drawn. More recently, a study by Robert Blust of the University of Hawai`i, entitled “Chamorro historical phonology” appeared in the journal Oceanic Linguistics in 2000.

This work provides a description of the sound system of CHamoru as it is currently spoken and gives evidence to show how each of the sounds has developed in CHamoru from Proto-Malayo-Polynesian (PMP), the parent of all the Austronesian languages outside of Taiwan, spoken around 2,000 BC, and which can be shown to probably be the immediate ancestor of CHamoru. (This parent language is also referred to in some descriptions as Proto-Extra-Formosan.)

Vowels

Proto-Malayo-Polynesian (PMP) had four vowels: i, u, a and e. In CHamoru the first three of these sounds remained unchanged, although in some positions in a word, “i” became “e”, and “u” became “o”. PMP “e” was not a sound like Chamorro “e”, but rather a sound pronounced similar to the vowel in English ‘hurt’, sometimes referred to as “schwa.”

In CHamoru, this sound also became “u” (and in some positions “o”). Examples: PMP Rebek > gupu ‘to fly’, beRas > pugas ‘husked rice’, tebuh > tupu ‘sugarcane, qatep > atof ‘roof’, paniki > fanihi ‘flying fox’, etc.

Consonants

Proto-Malayo-Polynesian consonants developed regularly in CHamoru as follows. (Not all PMP consonants are shown here and many details of development are left unstated; however they serve to indicate the regular nature of the sound changes that have taken place in the language from its proto-language.)

p > fpasu > fasucheekpaqit > fa’etsalty; bitter
t > ttelu > tuluthreemata > mataeye
c > sceŋceŋ > songsongstopper, plug
k > h (before a vowel)kutu > hutulouselaki > lahiman, male
k > (at the end of a word)tasik > tasiseaRebek > guputo fly
q > ‘ (glota)ma-qasiq > ma-’ase’mercifulRumaq > guma’house
b > pbulan > pulanmoonbabaq > papa’down, below
d > hdanum > hanomwatersida > sihathey, them
z > chzalan > chalanroad, pathquzan > uchanrain
j > ’pajay > fa’irice in the fieldlalej > lalo’housefly
g > gganas > ganasappetitegetus > gutossnap, break off
m >mlima > limafivedalem > halomin, into, inside, enter
n > nnepuq > nufo’scorpion fishpaniki > fanihiflying fox
ñ > ññamuk > ñamumosquitolaña > lañaoil (generic)
ŋ > ngdeŋeR > hungokheartuqelaŋ > to’langbone
s > ssusu > susubreastnusnus > nosnoscuttle fish, squid
R > gRebek > guputo flyzuRuq > chugo’sap, juice

Subgrouping Hypotheses

From the above it is clear that CHamoru is an Austronesian language. There are hundreds of words that probably directly reflect words that have been reconstructed to Proto-Malayo-Polynesian. Knowing this however doesn’t tell us who the closest relatives of CHamoru are, nor where the original CHamorus came from.

Subgrouping claims

There are at least four main claims regarding the subgrouping relationship of CHamoru. These are:

  • CHamoru is part of the Philippine family of languages.
  • CHamoru is most closely related to certain languages in Indonesia.
  • CHamoru is most closely related to some of the Austronesian languages in Taiwan.
  • CHamoru is not closely related to any other subgroups within the Austronesian language family.

Philippine connection

The first claim has appeared in various publications but probably is best known from the often-cited comment found in Topping’s Chamorro Reference Grammar (1973) that “Chamorro is a Philippine type language, and its closest linguistic relatives are probably Ilokano and Tagalog. This opinion is based on the many similarities in the grammatical structures of the languages…. It is quite possible that these similarities in the grammatical devices were borrowed from Filipinos with whom the Chamorro traded. However this is very unlikely.”

Topping’s claim was based, not on sound changes, but on the apparent similarity in the way verbs are formed in each of these languages. The so-called “focus system” of Philippine-type languages which appears also in some sentence types in CHamoru is not unique to these languages, it is found in the Austronesian languages in Taiwan, some of the Indonesian languages, and even in Malagasy spoken in Madagascar, and is reconstructible to very early stages of the Austronesian language family, so it cannot provide evidence of a close relationship between the Philippine languages and CHamoru.

There are a considerable number of words in CHamoru that are similar in form and meaning to words that are found in Philippine languages, but unless the CHamoru words show the sound changes that characterize its inherited vocabulary from Proto-Malayo-Polynesian, the words must be assumed to be borrowings, borrowed into CHamoru by Filipinos coming to live and work in the country.

Indonesian connection

The second claim has been argued by Erik Zobel, who attempts to relate both Palauan and CHamoru to a subgroup of Austronesian languages that he labels Nuclear Malayo-Polynesian, and which he claims consists of a number of the languages of Indonesia, mainland Southeast Asia, and the Oceanic languages, but which explicitly excludes the languages of the Philippines, northern Sulawesi and most of the languages of Borneo and Madagascar.

Zobel’s claim is based on some similarities in the grammars of CHamoru and Palauan with the languages in his Nuclear Malayo-Polynesian subgroup. This is an attractive hypothesis, because it suggests the possibility that the original migrants who eventually settled the Marianas left from the northern Sulawesi area and after establishing an intermediate settlement in Palau, moved north to populate the Marianas, but its claims have been challenged.

Other scholars addressed each of the claims made by Zobel, and showed that each of Zobel’s proposed innovations is also found in some Philippine languages. Reid argues that the similarities are not the result of innovations in a shared parent language, but were probably independent changes that have taken place in those languages in which they are found, or they are the result of contact between the languages. Moreover, as Blust has noted, there are no uniquely shared sound changes or lexical items as one would expect if the hypothesis was valid.

Taiwan connection

In 1991, Stanley Starosta and Louise Pagotto compared the syntactic features of CHamoru with Tsou, an Austronesian language spoken in Taiwan, and also with Tagalog in the Philippines, and Marshallese, a Micronesian language. They concluded that each of the shared syntactic features were probably inherited from Proto-Austronesian and did not provide any evidence for a close relationship of CHamoru with any one of these languages.

Subsequently, however, Starosta (in 1995) claimed that CHamoru, while sharing aspects of its grammar and word formation with the Taiwanese languages Rukai, Tsou, and Saaroa, does not use a number of affixes that he claims developed in other languages in Taiwan, and which are shared by other Austronesian languages outside of Taiwan. He thus claimed that CHamoru separated at a very early stage, sailing from Taiwan, before the development of those affixes and before the movement south that brought about the occupation of the Philippines and other areas now occupied by speakers of Austronesian languages.

Starosta’s hypotheses regarding the subgrouping of the Austronesian languages in Taiwan have not been widely accepted by linguists, and Reid (2002) throws doubt upon the validity of his claims about CHamoru, by showing that some of the affixes that he proposed had not been innovated until after the ancestors of the CHamoru had left Taiwan, do in fact appear in the language, either with shifted forms or functions.

Malayo-Polynesian connection

Blust states:

Chamorro is descended from a single language ancestral to all Austronesian languages outside of Taiwan. Within the Malayo-Polynesian group, Chamorro shows no close ties with any other language.

An examination of the sound changes that are found in native words in CHamoru reveals that a number of changes that occurred in the language of the first migrants to move out of Taiwan into the Philippines (Proto-Malayo-Polynesian), are found also in CHamoru.

For example, some of the sounds that were differentiated in Formosan languages were not distinguished in Proto-Malayo-Polynesian, neither are they distinguished in CHamoru. Some words that were innovated in Proto-Malayo-Polynesian are also found in CHamoru.

CHamoru reflects some changes in its grammar that took place in Proto-Malayo-Polynesian, such as the systematic use of the prefix man- with accompanying changes to the final nasal sound, depending on the initial consonant of the base to which it was attached, e.g., man- + gupu > manggupu ‘to fly (plural subject)’.

If in fact the first CHamoru migrants probably set sail from the Philippines, as linguists claim, why is that CHamoru is not most closely related to Philippine languages?

It is because the first CHamoru migrants left the Philippines before the Proto-Malayo-Polynesian language began to differentiate into the dialects which resulted in today’s Philippine languages. This claim is supported not only by the very early archaeological dates for first occupation in the Marianas, but also by the linguistic evidence. There are several words, for example, that have been reconstructed for Proto-Malayo-Polynesian, whose reflexes in Philippine languages show changes that are not found in CHamoru. For example PMP Rumaq ‘house’ (Chamorro guma’ ‘house’) is reflected in Philippine languages with the meaning ‘sheath’, and #siwa ‘nine’ (Old Chamorro sigwa) is reflected in Philippine languages as either siyam or siyaw ‘nine.’

Saipan settled first

Blust also notes that the reflexes in CHamoru of the Proto-Malayo-Polynesian directional terms “lahud” ‘towards the sea’ (in Saipan and Rota lagu ‘west’, in Guam ‘north’) and “daya” ‘towards the mountains’ (in Saipan and Rota haya ‘east’, in Guam ‘south’), imply that initial settlement in Saipan was from the west, while that in Guam was from the north.

This is consistent with an initial migration from the Philippines into Saipan, with subsequent migration south into Guam. That the migration was probably from the northern or central Philippines, rather than from the south is supported by the presence of a reflex of PMP “baRiuh” ‘typhoon’ in CHamoru pakyo ‘typhoon, storm, tropical cyclone’, since typhoons although a regular occurrence in Luzon and the central Philippines, are virtually unknown in Mindanao and areas south of the Philippines, excluding Sulawesi and Palau as possible places of origin.

Evidence for intensive contact

According to Blust, there is no evidence from either archaeology or the sound changes that have taken place to support a popular hypothesis (suggested also by Costenoble’s claims of “layers” of vocabulary) that the original CHamoru population was conquered by an in-migrating group of people who brought with them latte and rice agriculture and left a major imprint on the language. Blust does however provide a number of CHamoru terms whose origin can only be from one or more of the Micronesian languages.

Furthermore, Reid’s examination of the grammar of CHamoru suggests that there are several sets of parallel constructions, or syntactic doublets, suggesting the possibility that the language has undergone considerable change as a result of intensive language contact. One of these doublets is the set of possessive constructions, such as i gima’ ña si Rosa ‘Rosa’s house’ or i hagaña i rai ‘the king’s daughter’ versus i gima’ Rosa ‘the house of Rosa’ or i hagan rai ‘the daughter of the king’.

Another set are the existential possessive constructions, such as Guaha salape’hu ‘I have money’ or Taya’ salape’hu ‘I have no money’ versus Gai salape’ yo ‘I have money’ or Tai salape’ yo ‘I have no money.’ Different sets of transitive and intransitive constructions with the same or similar meanings, provide further support for this hypothesis.

Knowledge of rice cultivation

This does not support however the claim that the knowledge of rice agriculture was introduced as a result of language contact, nor is it possible to unambiguously locate the source of the language(s) that might have influenced CHamoru to this degree.

Terms for rice, rice cultivation and preparation have been reconstructed for Proto-Austronesian and must have been known by the first CHamoru migrants, since these terms, such as fa’i ‘rice in the field, rice plant’ (from PMP “pajay”), pugas ‘uncooked rice’ (from PMP “beRas”), and lusong ‘rice mortar’ (from PMP “lesuŋ”) clearly reflect the sound changes that characterize inherited, native vocabulary.

Scholars also know from archaeology and from linguistics that rice agriculture was a significant part of the culture of the earliest migrants to Taiwan from Southeast Asia, and was carried with them when they eventually set sail for the Philippines and islands to the south.

If the earliest CHamorus did not have rice agriculture for several thousand years, how were they able to maintain knowledge of the terminology associated with it? The terms would have been forgotten within a few generations. The fact that archaeologists have not yet found very early evidence for rice, only means that the sites where it may have existed have not yet been discovered.

The linguistic evidence then, supports the fact that the CHamoru language is a Malayo-Polynesian Isolate.

By Lawrence A. Reid, PhD

For further reading

Blust, Robert. “Chamorro Historical Phonology.” Oceanic Linguistics 39, no. 1 (2000): 83-122.

Reid, Lawrence A. “Morphosyntactic Evidence for the Position of Chamorro in the Austronesian Language Family.” In Collected Papers on Southeast Asian and Pacific Languages. Edited by Robert S. Bauer. Canberra: Pacific Linguistics, 2002.

Zobel, Erik. “The Position of Chamorro and Palauan in the Austronesian Family Tree: Evidence from Verb Morphosyntax.” In The History and Typology of Western Austronesian Voice Systems. Edited by Fay Wouk and Malcolm Ross. Canberra: Pacific Linguistics, 2002.