Excerpt from “The Haunted Lotus” by Australian-Hazara artist Khadim Ali (Source: Milani Gallery)

An objection that is commonly raised against Esperanto and other auxiliary languages is that they are “invented.” Somehow, being “invented” is assumed to give Esperanto a shady character: it’s just not natural. The problem with this view is that – in being invented – Esperanto is not unique. And I don’t just mean that there is also Klingon and Volapük. In fact, each and every language with a name is an invention. We may not always be able to identify the inventors – in fact the trick of the inventors of English, Chinese, German, Spanish and all the others – has been not to let themselves be identified as language inventors. Instead, they pose as teachers, priests, bureaucrats, academics, poets or scientists. The invention of major national languages such as these gets obscured by time (although Standard German with its origins in the 19^th century is not much older than Esperanto), and it is a rare opportunity to see a language invented before our own eyes.

Such an opportunity currently unfolds in Australia with the invention of the Hazaragi language. Late last year I was invited to attend the 2013 NSW Fair Trading Think Smart Multicultural Conference. Among the many important things I learnt at that conference was the discovery of a multilingual resource for renters in New South Wales. The video “Renting a home: a tenant’s guide to rights and responsibilities” is an excellent educational resource and it is available not only in English but, additionally, in 17 other community languages. What struck me was that three of these 17 languages were the same, as far as I am concerned: there is “Dari,” “Farsi” and “Hazaragi.” Isn’t it all Persian, I thought? I was aware that “Farsi” is often used for “Iranian Persian” and “Dari” for “Afghan Persian” but I had never encountered “Hazaragi” listed as a separate language before; it is usually treated as the Persian dialect spoken by the Hazara of Afghanistan. The Hazara are Shia Muslims of Mongol ancestry whose traditional homeland are the high mountains of central Afghanistan (Farr 2007).

So, I did some research and discovered that Hazaragi is a language that is currently being invented in Australia and linguists from around the world might wish to pay close attention how this process unfolds.

To begin with, it’s imperative to identify speaker numbers because you can’t have a “natural” language without a community of speakers – and remember I’m talking about concealed invention; not something as straightforward as Ludwik Zamenhof saying “an international auxiliary language is a great idea and I’m going to create one.” In order to achieve speaker numbers, the categories of the Australian national census had to be adapted a bit over the years, as a comparison of the category for “Persian” over five consecutive censuses shows: the 1991 Census had no category for Persian nor related varieties and they were all subsumed under “Asian Languages, not elsewhere included.” Reflecting growing immigration from Iran, by the next census in 1996, “Persian” had its own category, which remained unchanged in 2001. The 2006 Census saw a significant change to the category when the language label was changed to “Iranic languages” with three distinct subcategories: “Persian (excluding Dari),” “Dari” and “Other.” “Other” was defined to comprise “Iranic, not further defined,” “Kurdish,” “Pashto,” “Balochi” and “Iranic, not elsewhere classified.” (There is no need to write in and ask what the difference between “Iranic, not further defined” and “Iranic, not elsewhere classified” might be. I don’t know.) It was not until the 2011 Census that “Hazaragi” made its debut, when it was included in the “Iranic Languages, Other” category for the first time. The table visualizes the changes in category.

Another important aspect of instituting Hazaragi as a language in Australia is through the credentialing of interpreters. NAATI, the National Accreditation Authority for Translators and Interpreters, credentials Hazaragi paraprofessional interpreters through testing. On inquiry, I have learnt that NAATI decisions about recognizing a variety as a language are based on Australian Bureau of Statistics data (see above) and “voices from the community about the designations that they use for themselves.” In fact, it seems quite impossible to find out how the decision to accord language status to Hazaragi was made. Even so, NAATI has clear guidelines as to what is correct and incorrect Hazaragi in a testing situation:

NAATI acknowledges that there are regional variations/dialects of the Hazaragi language. However, due to strong cultural and identity connections there is a high level of mutual understandability between these regional dialects.

For the purposes of NAATI testing, a candidate will not be penalised for the dialect spoken as long as what is being said would be understood by an average Hazara person living in Hazaristan.

Candidates need to be aware that the Hazaragi language spoken by Hazaras in some locations, including the major cities in Afghanistan, has been heavily influenced by other languages of those cities and areas. Any use of ‘none’ [sic] Hazargai’ [sic] words when interpreting would be penalised. (NAATI Information Booklet)

This statement is a crucial step in the invention of the Hazaragi language. After the language has been given a name, it is being codified. Again, the process of invention is dissimulated: the language spoken in the mythical place of origin, Hazaristan (incidentally, there is also a little identity war going on over whether that region should be called “Hazarajat” or “Hazaristan,” the latter supposedly being “more modern”) is normalised whereas language use that shows traces of the influence by other locations, particularly cities, is penalized, presumably because someone got it into their head that such influence is “incorrect.”

This particular invention – Hazaragi as the language of rural Hazaristan – is rather baffling: from an Australian perspective, the language spoken by “an average Hazara person living in Hazaristan” is entirely irrelevant because even if such persons were to exist in Afghanistan, they do not in Australia. The past three decades or more have been an unmitigated disaster for Afghanistan and have produced the world’s largest refugee population. Contemporary Hazara society is characterised by constant migration:

Like most Afghan groups, the Hazāras fled in large numbers after the coup of April 1978 and the Soviet intervention in 1979. Most of them went to one of the neighboring countries of Afghanistan. Migrants and refugees have thus come to overlap and can hardly be distinguished from each other. Their movements follow various patterns: thousands of farmers from the Hazārajāt migrate every winter to work in coal mines near Quetta for a few months, while young men migrate for longer periods to Iran to take on menial jobs. During the last two decades, the Hazāras have formed very efficient migratory and economic networks, based on the dispersion of relatives in Iran, Pakistan, and Afghanistan. (Encyclopaedia Iranica)

Hazaragi has always been a contact variety – its main claim to distinction from Persian is the relatively higher number of Mongol loan words – and, in all likelihood, will continue to be a contact variety for a long time to come. It’s hard to see how inventing boundaries and a standard for this variety will do any good to anyone. Peter Mühlhäusler (2000) has an apt term for this kind of linguistics: segregational linguistics.

Farr, Grant. (2007). The Hazaras of Central Afghanistan. In B. Brower & B. R. Johnston (Eds.), Disappearing Peoples?: Indigenous Groups and Ethnic Minorities in South and Central Asia (pp. 153-168). Walnut Creek, CA: Left Coast Press.
Mühlhäusler, Peter (2000). Language Planning and Language Ecology Current issues in language planning, 1 (3), 306-362 DOI: 10.1080/14664200008668011

The challenges of providing equitable services in linguistically diverse societies

A few years ago, emergency service provision to speakers of languages other than English in Australia came under scrutiny when an Afghan woman tried to call the police but did not receive any assistance a few days before she was murdered by her husband. The domestic violence victim called 000 – the national telephone emergency service – twice but, in both cases, she hung up before the operator could bring a telephone interpreter into the conversation. The trial judge was scathing of multilingual service provision in her sentencing:

Then on 1 November 2007 your wife made two calls to triple 000 in the early hours of the morning seeking police assistance. It is most unfortunate and an indictment on our society that no assistance was forthcoming, as a result of a very disappointing reaction by the telephone operator, to your wife’s inability to speak English, in anything other than a broken English manner. (“R V Azizi,” 2010, p. §11)

Providing emergency services in linguistically heterogeneous environments is not an easy task. In Australia, as elsewhere, calls from callers who cannot make themselves understood in English are re-routed to the national telephone interpreter service. Negotiating for an interpreter service and then waiting for the interpreter to come online in an emergency is time-consuming and stressful. As Raymond (2014) found in a study of Spanish calls to the US emergency telephone service, sending a call through to the telephone interpreting service more than doubled the response time and thus was often resisted by the operator because that additional time could make the difference between life and death in an acute emergency.

Given the time it takes to re-route an emergency call through the telephone interpreting service, it is perhaps not surprising that the domestic violence victim mentioned above became impatient or was forced to abandon her call for help before she could actually communicate with the police. What is surprising is that no attempt was made to call her back and no police car was dispatched to her place of residence despite her two failed attempts to call emergency services.

Domestic violence is a well-recognized problem in Australia and there is a network of formal support services available to women who find themselves in danger. However, these formal support services operate exclusively in English and gaining access in another language remains difficult. As an examination of the barriers faced by migrant women in accessing legal services in NSW put it, persons who do not speak English well in Australia are “a long way from equal.” Cases like these challenge us to rethink linguistic arrangements in linguistically diverse contexts in order to make access to emergency services – and social services more generally – more equitable. It may be tempting to think that this is too tall an order and that ensuring equitable access in any language other than the nationally dominant language is too difficult.

Not so! In the above-mentioned study of Spanish-language calls in the USA, transfer of a Spanish-speaking caller to the telephone interpreting service almost doubled the response time and thus was highly inefficient. However, the situation was different in cases where bilingual operators were available in the same call center. In such cases, a Spanish emergency call was immediately transferred to a bilingual operator and the response time was virtually unaffected. So, a way to provide effective multilingual services would be to hire multilingual telephone operators who can handle calls in the dominant language and one or more other languages.

Now here is the rub: in the city where the researcher recorded the bulk of his data no single bilingual operator who spoke Spanish was on the staff. When one knows about the demographic profile of that city, the absence of English-Spanish bilingual emergency telephone operators becomes almost unbelievable: the city where the research took place is located in the US Southwest and at the time of data collection in 2010 more than 50% of the city’s population were of Hispanic/Latino origin. This information turns the assumption that monolingual emergency services are natural and normal on its head: only employing monolingual English-speaking emergency telephone operators in such a bilingual context almost seems perversely designed to prevent fair and equitable access. I am not for a moment suggesting that monolingual emergency services are the result of some sort of conspiracy. They are not. They are an expression of our collective failure of imagination: a failure to recognize that linguistic diversity poses an equity problem and a failure to imagine that we can change our social linguistic arrangements in ways that make them more equitable and just.

Raymond, Chase Wesley (2014). Negotiating Entitlement to Language: Calling 911 without English
Language in Society, 43 (1), 33-59 DOI: 10.1017/S0047404513000869

I can only speak Pashto

ABC News yesterday broke a whistleblower report that US army interpreters deployed in Afghanistan often don’t speak the local languages. ABC News reports that up to a quarter of the interpreters hired by a private provider, Mission Essential Personnel, failed basic language tests.

Click on the picture, to watch one such interpreting episode unfold: a Pashto-speaking villager in effect offers to cooperate with US troops but the interpreter makes up an entirely different scenario with the Americans eventually walking away and cursing the villager. No wonder they don’t get anywhere winning the hearts and minds of the locals!

A Christian Science Monitor report has some examples of the damage caused by incompetent interpreters: in one case, misinterpreted directions resulted in a misdirected mortar attack on the wrong spot. The entire livestock of a village was killed in that attack; US troops paid compensation to the villagers. In another example, a request for shooting illumination flares was misinterpreted as a request for a mortar attack, which resulted in an unspecified number of casualties.

The reasons for these disasters are manifold: to begin with, war interpreting is a lucrative business both for the interpreters on the ground but even more so for corporations such as Mission Essential Personnel. Second, Afghanistan is a multilingual country and apparently someone hired as a Dari interpreter may well be then assigned to interpret in a Pashto- or Baluchi-speaking part of the country. Third, army interpreters are often uneducated and inexperienced young men as the more senior interpreters opt for more secure employment with the UN or NGOs in Kabul. One of the interpreters interviewed for the Christian Science Monitor report, for instance, tells how he learnt English by selling cigarettes to soldiers outside Bagram Airbase. He was only 16 when someone asked him whether he spoke Dari and Pashto. He said “yes” and, voila, he had a new job as army interpreter.

Would anyone have given him a job as, say, an accountant? “Hey, are you good at maths?” “Yes!” “Beaut! I’ve got a job for you as an army account.” Wouldn’t happen because everyone understands the need for accountants to be properly trained, qualified and to have substantial experience for a high-stakes role. By contrast, decision makers in the US army are seemingly so naive about language and communication skills that they think nothing of putting the lives of civilians and soldiers, indeed the entire outcome of the operation, into the hands of untrained, unqualified and inexperienced interpreters. Just another reason to Rethink Afghanistan!