The southernmost settlements of America are shared between two countries — Argentina and Chile. Nestled in the island called Isla Grande de Tierra del Fuego, translating to “Big Island of the Land of Fire”, this island just south of continental South America is home to the settlements of interest today. Among these, are three settlements that collectively claim to be the world’s southernmost city, with all of them located south of 53°S and largely sharing the same climate. The largest and northernmost among which is Punta Arenas in Chile, followed by Ushuaia in Argentina, and lastly, Puerto Williams in Chile. Depending on how you define what a city is, these three settlements do have some claim to that very title. But for the purposes of this essay, we will use the term ‘settlement’ instead.

Puerto Williams is home to around 9000 people, but has a history of settlement by the indigenous Yaghan people, who settled the southern regions of Isla Grande de Tierra del Fuego to escape conflict in the north. It is their language, the Yaghan language, which gave the Argentinian settlement of Ushuaia its name, which could be split into the constituent words ush and waia to mean ‘deep bay’.

The Yaghan language is also the language where we could find what is purported the ‘most succinct word’, or among the most ‘untranslatable’ words. It is called mamihlapinatapai, split into elements mam + ihlapi + n + ata + apai. If we take these elements and translate them literally, the word would mean ‘to make each other to be at a loss of what to do next’, or perhaps in short, ‘to make each other feel awkward’.

But after factoring in nuances, there are several translations proposed. The common element among these is the part where both parties referred to by the word are ‘unwilling to initiate’, or ‘unwilling to do [something]’. Other elements are often described as something to do with two people looking at each other, with the expectation that the other person would do or start something. But the issue is if this exact word has actually been used in Yaghan speech. One of the first linguists who worked on the language, Thomas Bridges, only recorded the word ihlapi, meaning ‘confused’ or ‘uncomfortable’. A century later, another linguist by the name of Yoram Meroz proposed derivations from that root word, by attaching affixes denoting grammatical functions to that word. Others have also noted Bridges’ tendency to use illustrative examples of a Yaghan word to inform the reader about that word’s meaning.

So, we have been using the word Yaghan a lot so far. But the language goes by several more names, such as Yagán, Yámana, Háusi Kúta, Yágankuta, and Yahgan (yes, the ‘g’ and ‘h’ are swapped in this one). There does not seem to be a uniform way of referring to the people nor the language, but for purposes of this introduction, we will use the term Yaghan, one of the terms from which the ISO 639-3 code, yag, is derived. It is a language isolate, although there is some influence from languages like the Kawésqar language, as the Yaghan people regularly interacted with the Kawésqar people.

Today, there are no native speakers of Yaghan, with the passing of the last known native speaker, Cristina Calderón, in 2022. But this is not the end of the Yaghan people and culture though. There are around a thousand Yaghan people today living in Chile and Argentina, but most of them speak Spanish. Nevertheless, we have some resources to work with, compiled from studies on the Yaghan language over the past hundred years or so.

Some of the earliest linguists who have worked on the Yaghan language include the aforementioned Bridges in the 19th century, Perla Golbert de Goodbar in the 1970s, and Ana Maria Guerra Eissmann in the 1990s. While Bridges’ work was published in English, many of these resources on the Yaghan language are in Spanish. But for a more comprehensive summary of the Yaghan language in English, you could try the 2004 publication titled ‘The Languages of the Andes’ by the linguists Willem F. H. Adelaar and Pieter C. Muysken.

There is a bit of disagreement on whether or not Yaghan has 6 or 7 phonemic vowels, which largely pertains to the distinction or lack thereof between the vowels [ɑ] and [ə]. Another disagreement seems to pertain to whether [æ] is an allophone of some other vowel. Adelaar and Muysken seems to use the analyses done by Golbert de Goodbar in 1977 and Poblete and Salas in 1999, which recognises a 6-vowel system, [ɑ] and [ə] being allophones, and suggested that [æ] was not a phonemic vowel. As such, the 6 Yaghan vowels are [i], [u], [e], [ə], [o], and [u]. Depending on the source, Yaghan is recognised to have 15-18 consonants, though Bridges’ analysis a century prior recognised a few more sounds.

From how the word mamihlapinatapai is formed, we can see that Yaghan has a bunch of prefixes and suffixes that attach to a root word, like a noun or a verb, to reflect various grammatical functions. This heavy use of suffixes and prefixes allows for a fairly more flexible word order in Yaghan, and slightly complicate the basic word order for the language.

Firstly, we will look at the noun. What I first noticed when reading the glosses in Adelaar and Muysken’s brief sketch was the kinds of words which are loaned from other languages like English. Words like ‘leaf’, ‘glass’, and ‘pocket’ are loaned from English, which seemed really peculiar. It turns out that the Yaghan people live in a place with few land resources, and never quite settled further inland. This sort of explains the lack of Yaghan terms that have to do with land. Instead, the Yaghan tended to venture seawards, with more terms dedicated to sea life and birds. There are more kinship and body part terms as well.

The boundaries between nouns and adjectives can be quite blurred, but they are always distinct from verbs. To make verbs from adjectives, for example, you could do so in several ways. To ‘be’ in a certain state, you would lengthen the vowel at the end of the root, and attach a -na. To develop towards a certain state, this would be further suffixed with -ata, resulting in the lengthening of the vowel at the end of the root, plus the -nata suffix. Using the similar pattern, but with a -nuhrka:taka suffix makes it ‘incrementally become’ a state.

Verbs have a bunch of compounding and derivational affixes. Yaghan follows an ergative-absolutive alignment, and hence for verbs that carry a direct object, the verb would agree with the object. In Adelaar and Muysken’s book, he noted Bridges’ account of Yaghan verbs, and remarked how different the verbs agreed with plural objects, which were substantially different from singular and dual objects. This also applies for verbs that do not take direct objects, in which case they agree with the subject. Take a look at the following to make a comparison:

Singular	Dual	Plural	EN
ta:gu:	ta:gu:pay	yatu:	to give [object]
ata	ata:pay	tu:mi:na	to take [object]
apəna	apəna:pay	ma:maya	[subject] die(s)
ka:taka	ka:taka:pay	u:tušu:	[subject] go(es) on foot

Because of the compounding affixes that can be attached to a root verb, Yaghan seems to have a rather large lexicon of verbs. This shares similarities with some languages of North America such as the Algonquian languages, and the neighbouring language once spoken in Isla Grande de Tierra del Fuego, Selk’nam. Verbs of motion may be prefixed using directional prefixes, such as outward ma:n– or mana-, and eastward mət– or məta-. These prefixes may also be used as suffixes when attached to a noun, however. Unlike other verbs, modal verbs such as ‘can’ káy are not inflected for person and tense. For more information about the Yaghan language, I really recommend checking out Adelaar and Muysken’s 2004 publication. It is in English, and while it contains linguistics terms, the glosses are helpful in dissecting Yaghan parts of speech.

Despite the passing of the last known native speaker of the Yaghan language, there is some optimism to be had. Currently, the Endangered Languages Project classifies Yaghan not as extinct, but as dormant. This means that there are revitalisation efforts to revive the language for the Yaghan people. Cristina Calderón’s legacy also lives on, as some Yaghan songs, tales, and legends are published in a book titled, in Yaghan, Hai Kur Mamašu Shis, or “I want to tell you a story”. This ensures the preservation of Yaghan culture, and offers more avenues, alongside linguistic research, for revitalisation of Yaghan to take place.

Currently, Chile seems to have more prominent projects dedicated to Yaghan revitalisation, with an official orthography, and a publication of an illustrated Yaghan dictionary in 2010 which you can find here. There are also plans to make the surrounding area around Puerto Williams to be the centre for language revitalisation, and teach the younger Yaghan the language which was recently ‘lost’. Through cooperation between linguists and local communities, there is certainly optimism in bringing the Yaghan language back to life.

Further reading

Adelaar, W.F.H. and Muysken, P.C. (2004) The Languages of the Andes. Cambridge: Cambridge University Press (Cambridge Language Surveys).

Grenoble, L.A. and Whaley, L.J. (2002) ‘What does Yaghan have to do with digital technology?’, Linguistic Discovery, 1(1). doi:10.1349/ps1.1537-0852.a.101.