tashu duset sekar

Helen DeWitt

Consider the following excerpts from a couple of language aptitude tests:

Test 1

The questions in this section are based on an invented language, called Dobla. Read each group of examples carefully, paying particular attention to different forms of words and working out what information they convey (just as in English there are differences between e.g. cat and cats, or beckon and beckoned). Word order in Dobla is different from that of English and is not entirely fixed; it is not a reliable guide to the meaning of sentences. Note also that Dobla has nothing corresponding to English the and a(n), so that tine can mean either ‘the maid’ or ‘a maid’. You are advised to work through the questions in this section in the order in which they are given, as the later ones may presuppose information or vocabulary supplied in the earlier examples.

tashu duset sekar

‘The diplomat seduces the daughter.’

tine betsut vardar

‘The maid helps the valet.’

betsu tinet sirehar

‘Does the valet love the maid?’

claru bichut sudar

‘The earl consults the butler.’

vardehar bichu kochet

‘Does the butler help the cook?’

pante sirar tomut

‘The scullery-maid loves the footman.’

rokar elede duset

‘The countess summons the daughter.’

clarut tikehar mage

‘Does the dowager rebuke the earl?’

Give the meaning of:

tomu sekehar tashut

pantet tikar koche

Translate into Dobla:

‘Does the daughter consult the dowager?’

Test 2

Rubric as above, the invented language is Kalaamfaadi.

Felhom ghrabop karhyd

The farmer hates the crow

Milkyd felhom rassop

The farmer owns a gun

Tayrom rassop karhyd

The bird hates the gun

Sadyd tayrop sabyom

The boy helps the bird

Bussyd felhop ghrabom

The crow sees the farmer

Extayrop felhom talkyd

The farmer shoots the birds

Exsabyom extalkyd felhop

The boys shoot the farmer

Give the meaning of:

Exfellop sadyd rassom

Extayrom exsabyop exbussyd

Translate into Kalaamfaadi: ‘The crows own guns.’

In Test 1, the invented language questions are part of a longer paper; two other sections, each worth 25 points, give examples from existing languages (Hungarian, Czech, English) and invite the candidate to extrapolate, comment. The candidate has an hour to complete the test. Test 2 involves only invented language questions; the candidate has 30 minutes.

Test 1 is an extract from the Classics Language Aptitude Test, a one-hour test taken by Oxford applicants who want to study classics, either on its own or as part of a joint course, but have neither Latin nor Greek A-Level. Test 2 is an extract from the Oriental Languages Aptitude Test, a 30-minute test for Oxford applicants in Oriental Studies, who will need to learn Arabic, Persian, Hebrew or Turkish from scratch. The Japanese Studies website suggests that it is also necessary for would-be students of Japanese.

The Oriental Studies website says explicitly that the test cannot disqualify a candidate from being invited for interview; it serves only to give interviewers more information on how well the candidate is likely to cope with a demanding part of the course. The Classics website implies that this is the case (performance on the test is not included among criteria for interview).

The tests obviously give interviewers some sense of how well a student is likely to cope with something he or she has not tried before. (The modern languages most taught in schools are not much of a guide: they are not radically different from English, and they do not have to be learned very fast.) But surely if a student has such an aptitude, it is much more important for the student to be aware of the aptitude, and to be aware of this as early as possible.

Latin and Greek are predominantly offered at private schools. So the languages are strongly associated with class. Arabic, Hebrew and so on are not so strongly associated with class, but there is even less chance to study them in the school system. Students with no exposure to these languages have no reason either to think they might be good at them, or that anything in their background might signal promise in the admissions process. But a very large number of schoolchildren and their parents are likely to have been exposed to one or more fictional languages in popular culture. Elvish and Dwarvish, Klingon, Dothraki, Valyrian, quasi-Latin spells – these are part of the charm of The Lord of the Rings, Star Trek, Game of Thrones, Harry Potter, and the number of persons who have warmed to one or more of them is not small.

The sort of test offered above is one a very large number of students might easily manage, not only after A-level, but at a much younger age. The sort of mind that likes puzzles might well cope with it at eight or nine. The sort of mind that likes glamorous languages might start with ‘oculos reparo’ at eight, decipher Tolkien’s linguistic appendices in their early teens, and perhaps (with tolerant parents) try to analyse some Old Valyrian at 15 or 16. (Valar morghulis: ‘All men must die.’ Valar dohaeris: ‘All men must serve.’ Interesting…)

But none of this would give a student, parents or even a teacher any reason to think the student had precisely the aptitude looked for by certain Oxford faculties. Nor would it magically convey that, with languages more challenging than those normally taught in schools, the whole structure of GCSE and A-level can be dispensed with: the student can start from scratch and be rapidly brought to a high standard. Unlike Sudoku booklets, the tests are not sold by newsagents; unlike the novels of Rowling and Tolkien, they are not sold at WH Smith and Tesco. A student would only know if he or she had at least contemplated both applying to Oxford and studying one of the relevant languages in the first place. A very large number of students might quite reasonably never think to look; the test is offered only to an artificially small sample.

Oxford can only choose among those who apply. Any subject not taught at school requires a leap of faith on the part of the candidate; one involving an apparently daunting language requires a much greater leap of faith. But it’s entirely possible that the kind of intellectual agility such languages call for is a hidden strength in many students who don’t know they have it. The thing that looks most intimidating might be the thing that should inspire confidence.

I don’t mean to make a fetish of Oxford or of languages. I do think, though, that it would matter both to schoolchildren and their parents, particularly those outside the independent system, to know they could hack a Language Aptitude Test devised by Oxford for the courses likeliest to look esoteric. Britain has such a mania for testing, and the results of these tests colour the expectations of students, parents and teachers – and yet a child in disgrace for poor performance at Key Stage 5 could be the next Michael Ventris, the next Champollion. I think they should be told.


  • 3 September 2015 at 10:28am
    Phil Edwards says:
    Hearty agreement. Not everyone has the language thing (or keeps it past early childhood) - but for those of us who do, what a terrible shame not to get a chance ot exercise it.

    I was never more jealous of my son than when he entered the Linguistics Olympiad, which consists entirely of imaginary-language puzzles like these - it looked like an awful lot of fun (it's not open to adult entrants, sadly). He did French and Chinese to GCSE and is now studying Mathematics; his French is, selon moi, tout a fait affreux, but he's travelled around China on his own, so some of that clearly stuck.

  • 4 September 2015 at 5:53pm
    Timothy Rogers says:
    A nice article by Helen DeWitt on a very interesting topic. I went through one of these imaginary language tests (a sub-category of code-breaking or code-learning, I’m guessing) as part of a series of tests administered by the US Army back in the mid-1960s. Depending on your former educational level, you were eligible for taking the ALAT (Army Language Aptitude Test), which was very similar in its materials to the example of Dobla given in DeWitt’s piece. This applied to both officers and enlisted men (I was one of the latter). I don’t think conscripts were eligible, simply for the reason that the programs were fairly long and the Army only had their services for two years, so they did not wish to invest time and money in such a candidate. Basic training and “Advanced Individual Training” took up something like nine months, leaving them 15 months for their overseas service (with a one-year limit on service in Vietnam).

    I cannot remember the exact scoring scale, but for purposes of explanation here, say it varied from zero to fifty (50 being a perfect score). Admission into specific language training programs depended on having attained a specific level on the test. For example, Chinese and several other Oriental languages were considered to be the most difficult, requiring a minimal score of 40 out of 50. Russian and other Slavic languages required something like a score of 35, German and French 30, the “easier” Romance languages (e.g., Italian, Spanish) a score of 25, etc.
    I had studied German in high school and college and was fairly fluent as a result of having spent three months in the old DBR as part of a summertime exchange program that placed students in jobs forcing them to use the local language. I seemed to have a knack for languages at the time, which came to the fore at about the age of 16. So I did well on the ALAT – unfortunately this qualified me for the Vietnamese language training program, which I was in for 47 weeks before being posted to Vietnam (too smart for my own good, no doubt).

    With respect to Vietnamese, the shortest course ran for 12 weeks (a bare toe-dipping), and the longest, intended for those in the ASA (Army Security Agency, i.e., “electronic intelligence” involving eavesdropping and radio intercepts) ran to 74 weeks (those enlistees had a four-year contract; most others had a three-year contract). Because the manpower requirements for the Army in Vietnam ballooned so rapidly, they slackened their ALAT standards quite a bit in order to show that they were training a sufficient number of people in the language of our ally (the RVN government), so many people went through the program without ever learning enough of the language to be actually useful. Because it is both monosyllabic and tonal, it's not an easy language for Westerners, though reading proficiency is not so difficult for people with a general aptitude for language.

    As to Dobla, it’s obviously the language of a very hierarchical society, perhaps lacking a middle-class.

  • 8 September 2015 at 4:00pm
    Mark Sponge says:
    Somebody check me?

    1. Does the footman seduce the diplomat?
    2. The cook rebukes the scullery maid.
    3. Duse sudehar maget

    1. A gun helps the farmers.
    2. The birds see the boys.
    3. Exghrabom exmilkyd exrassop.

    • 9 September 2015 at 1:46am
      zeugmasyllepsis says: @ Mark Sponge
      Those were my answers as well, modulo word order.

  • 8 September 2015 at 6:28pm
    juji says:
    I check the Dobla

    tomu sekehar tashut = Does the footman seduce the diplomat?
    pantet tikar koche = The cook rebukes the scullery-maid
    ‘Does the daughter consult the dowager?’ = Duse sudehar maget

    Does the student seduce the examiner?