138 comments

[ 3.6 ms ] story [ 213 ms ] thread
I got Ebonics. This quiz is whack.
Ah see what you did dere.

For english dialect I got:

1. Singaporean 2. US Black Vernacular / Ebonics 3. American (Standard)

and for native language I got:

1. Swedish 2. Norwegian 3. English

I'm 100% native English

> I'm 100% native English

Same here regarding habitat (Gloucester), my results were:

Dialect: 0. English 1. Welsh 2. South African

Native: 0. English 1. Finnish 2. Romanian

But my mother is Welsh, and I lived in Cardiff for three years. I'd say that is fairly accurate.

(comment deleted)
Wait, does it actually say "Ebonics" instead of "AAVE" or something like that? No linguist uses that word.
"US Black Vernacular / Ebonics" is the exact phrase used.

Which I also received.

I got 1. Singaporean English 2. US Black Vernacular/Ebonics 3. American English (standard). My dialect is standard Midwestern. (Or if you want to get really technical [1], Inland North.)

[1]: http://aschmann.net/AmEng/

Same here, though I am an immigrant(been living in NYC), nobody has ever told me that I speak Ebonics.
A surprising number of immigrant groups pick up elements of Ebonics in their English dialects. In my area non-Black groups with notable dialectal markers typical of Ebonics seem to include Vietnamese, Koreans, Salvadorians and various Arab groups (not as common).
It seems that the distance between American English and AAVE is very small, which is natural I guess.

I got 1. Ebonics 2. New Zealand 3. English (England) for dialect and 1. Turkish 2. German 3. Finnish for native language. I'm native Chinese currently living in California.

That's right. In our data, Standard American, Canadian, & AAVE are all pretty similar. Partly that's a function of the questions we ask. But partly that's because there's a clear divide between the north american dialects and the UK-based ones.
Same

dialects 1. US Black Vernacular / Ebonics 2. Singaporean 3. American (Standard)

This is really off! My english is a mix between various London accents. However, since I often hang out with foreigners, I may be more acceptant of sentences with weird grammar. That would explain the Ebonics.

native language 1. Italian 2. Chinese 3. Dutch

Although I hangout with Chinese people, I don't speak any of those languages. Plus, it would be good for a language study test to differentiate Cantonese, Mandarin...

The model doesn't do a good job of telling apart the various Chinese languages (yet), so we aren't including that distinction in the feedback (yet).
I just took the test a second time, and I'm getting the exact same results, appart from my native languages which changed to: 1. Italian 2. Dutch 3. English

Well, that must be it.

Good point: "I may be more acceptant of sentences with weird grammar."
At the end it says: "Calculating score... Meanwhile, please answer these questions."

At first I was like, "They're kicking off a deferred job for that?"

And then I was like, "Oh, they just want me to answer questions."

Is there really a significant population that self identifies as speaking Ebonics?

That whole methodology seems rife with biases. The instructions are in standard american english. The pictures are confusing. The input controls are broken on mobile. I reject written sentences ad ungrammatical even though I hear and speak them fine.

Maybe not Ebonics, but probably AAVE. It's a pretty significant dialect in North America.
The instructions ask you to choose what feels right, not what you learnt at school to be grammatical. So you should expect poor results if you rejected right-feeling sentences.
I got Ebonics, and it guessed that I am Vietnamese or Finnish. I consider myself to speak standard American English, and I am a Korean. Not enough training data? Is that why it's on HN???
No idea why Finnish, but the Korean population near where I live often speak with notable dialectal markers indicative of Ebonics. I think it comes from the relative popularity of hip-hop culture in Korean popular culture.

This interview is a great example. https://www.youtube.com/watch?v=aeb-PJJN9sk

Well, plus I actually grew up in the states!
>Up the audience's expectations, the critics built.

I just got that sentence on asking if it were grammatical or not. I believe that is referred to as yoda grammar?

>Our top three guesses for your English dialect: >1. Canadian >2. American (Standard) >3. US Black Vernacular / Ebonics

Number 1 is correct. However I would admit to speaking american, and I did listen to a lot of rap growing up, so ebonics might be reasonable? But I am a white kid from the rich suburbs, so it's a bit odd but possible I suppose.

One of my favourite things to do in online forums such as Stack Overflow, where there are many non-native English speakers, is to determine where a person is from based on which grammatical errors they make. It’s more subtle than accent or other cues, but it is remarkably consistent.
I followed Andrew Ng's Machine Learning course on Coursera. The course itself was great, but halfway through I realized he had to be from Malaysia or Singapore, because he started talking about "alphabets" instead of "letters".
Yeah, besides word choice (“I have a doubt” = “I have a question” in Indian English) you also have more interesting things like discourse deixis—“consider that code: …” instead of “consider this code: …”—or how uncertainty is expressed—“some” instead of “a” for example shows up in Daft Punk’s “Something About Us”.
Interesting! "I have a doubt" is also a common mistake for us Spanish speakers. This is because for us "tengo una duda" is an acceptable way of saying "I have a question".
German here. It felt _really_ weird when my American boss asked me whether I have a doubt. Yes, most people in his team are from India.
As a former (non-native) resident of Singapore, that's one of those features of the local lingo that makes me shudder every time I see it, but it's probably already crossed the line from "mistake" to "accepted local evolution". It certainly is well entrenched, as I've seen it in bank letters about choosing good passwords, and it's in this university policy as well:

Have both upper and lower case alphabets

http://iits.smu.edu.sg/cyber-security/password-management

Quite a few other telltale Singlish features in there as well, eg. the use of "staff" to refer to a single employee and using "credential" instead of "credentials". Both of these are rooted in how Chinese and Malay lack a grammatical plural.

I find it really easy to tell people who are native german speakers, based on learning german for years, and having had grammar drilled into me. I dont think there's been a time when I've picked up on it, and been wrong
Hm, curious. What would be tell-tale signs of a native German speaker? :)
I've noticed some Germans tend to mix up the present perfect and the infinitive, since they are the same in German.

For example, "I eat" vs "I'm eating" vs "I'm going to eat".

Online they often combine two-word compound nouns into one word.
My old boss was once setting up a patient when he asked where she was from, "as I have a good ear for accents, but I can't tell if you're from Rhodesia or South Africa". Her response was "It is good - halfway through my childhood, we moved from Rhodesia to South Africa"...
Would love to see more details as to WHY I got the suggested responses. As a prescribed grammarian, born and raised in the northeast US, I was definitely surprised to see Canadian and ebonics. Assuming not enough training data yet...?
When we're done with the project, we'll put up a final version that actually gives you a breakdown of why you get the score you do, what parts of the world certain grammar rules are particularly common -- all the things people (including me!) want to know.

We're actually getting some of that up bit-by-bit as data comes in (you can watch for it on our site blog). Thank you to everyone who has been participating so far.

Error message: > Our apologies. Your results didn't transfer correctly to this page. Don't worry -- your data was saved. If you've forgotten what your results were, you can email gameswithwords@gmail.com and tell us your age, country, and what time you finished the task, we may be able to figure out what your results were. We apologize for the inconvenience.

I see - the test writers must be Canadian...

edit: and since you're here, that message occurred in http://www.gameswithwords.org/WhichEnglish/done.php# after the result had been displayed, after which NoScript or AdBlockPlus interfered when I clicked on the question mark.

Prescribed English grammar often has little to do with the actual vernacular. I've seen many prescribed "grammatical" sentences that are woefully awkward to my very native English speaking ears, while the "ungrammatical" but more common form feels much more comfortable.
It didn't catch my northeastern either, suggested I was native dutch :/
I'm Brazilian, and their top guess for native language was Portuguese. Good work!
I find myself thinking of scenarios where most choices would be correct. I play ____ the soccer team.

Pleased to meet you. I am an active sporty guy and 'I play in the soccer team'.

Pleased to meet you. I am a tick, a playful little parasite. I like to jump from person to person in small groups. Today, 'I play on the soccer team'.

Pleased to meet you. I am a playful infectious disease. 'I play inside the soccer team'.

Pleased to meet you, I am Lot's son. As you may have learned in a famous bible story, my mother turned into a salt statue while taking a casual glance at the destruction of Sodom. Less known is that she belonged to a soccer team and the whole team also turned to stone while in the middle of an important game. Today, the team has become a famous landmark and the grounds for a soccer school to which I belong. 'I play at the soccer team'.

This has to be best best comment I've seen about the project yet! You mind if I use this as an example for my students? (I mean, I will anyway. I just want to know if you mind.)
Aww, thanks :) Go ahead.
> (I mean, I will anyway. I just want to know if you mind.)

That's an excellent way to ask for permission while being clear that you don't need permission.

I'm always wary of asking for permission in situations where it's not absolutely needed. What if the person says no? Then if I go ahead and do it anyway, I look like a jerk.

I'll have to remember the way you did it.

So if it doesn't list English in the top three native language guesses, does that mean my English is very ungrammatical, or merely that it's ungrammatical in a particular way? All three languages were Scandinavian but I have no connection there.

The NY Times had a similar quiz that attempted to predict your region and, IIRC, predicted mine fairly closely. I didn't get such accurate results this time so I'm wondering if my grammar is slipping ...

The #1 guess for me was Norweigian, while English was #2. I've spent all my life in the American South.
The way I speak...

Pleased to meet you. I am an active sporty guy and 'I play on the soccer team'.

Also, the last variation, would be 'I play at The Soccer Team'. Landmarks are (usually) capitalized.

True. I was thinking of some way not to capitalize it, any ideas?
I'm a musician, and I ridicule the athletic group I hate the most using my music. 'I play at the soccer team'.
Meld Apple's iProducts with Google Play, then launch it at a team's media event: iPlay at the soccer team.
I thought the same, then decided to answer the questions with the assumption that I didn't require a tortured situtation to make something grammatical. There's always an edge case, but the quiz was looking for common use.

Also, where I am, a sporty person can either play 'in' or 'on' the soccer team, no parasite required.

I wouldn't say any of those options.

I'm British and I would play for the soccer team (well the football team, but I'm not going to quibble about that).

You can also play with the soccer team. Which team do you play with at the university? I play with the soccer team. (rather than the rugby or netball or basketball or whatever teams)

And yes, they should have used 'football', and let each locality think they meant a different sport :)

I call both Soccer and AFL Football, and call Rugby, Footy. I only ever talk about AFL with my mother, Soccer with friends and Rugby with no one.

It can be really confusing when all three are somewhat popular in one country.

League or Union?

I'd call rugby league "footy", but rugby union will always be "ruggers".

Maybe it's a regional thing (I'm from the East Midlands), but playing with the team has a different connotation from playing for the team to me - although both are valid sentences.

Playing for the team suggests that I'm part of that team. Playing with the team would say to me that I'm not part of them, but I've been co-opted for a game or something like that.

Oddly, I would say "playing with the orchestra" rather than "playing for ...".

This is really cool, I didn't know that so many of those sentences might be considered valid.

It got my country right with it's second guess (New Zealand), which is pretty good considering what a small sample of kiwis it must have.

Ticking both "she'll be right" and "it'll be OK" was probably the big giveaway.
Top 3 guesses:

   1. American (Standard)
   2. Singaporean
   3. Canadian
(#1 is correct, #3 is close)

Native Language

   1. English
   2. Norwegian
   3. Chinese
(#1 is correct, interesting about the other 2)
Mines are as below,

Our top three guesses for your English dialect:

    1. Singaporean
    2. South African
    3. New Zealand
(None of these are correct)

Our top three guesses for your native (first) language:

    1. Arabic
    2. Chinese
    3. Russian
(Only 2 is true, I would also be interested to see how does they rank the orders)
> I told Sally I was worried about the exam. She said, "Don't worry. ____________"

> She'll be right!

I feel like this question is cheating. This could be the only question in the survey and it would be able to identify Australians.

She'll be apples when she discovers this.
I clicked "She'll be right!" just to test if this was true and none of the results guess me to be Australian. I had originally thought the same as you did but was pleasantly surprised.
I didn't pick that option and it still put my number 2 dialect as Australian.
Why would Australians say this? Is it because for them an exam is a "she"? (disclaimer: not a native English speaker myself, though the test guessed "American English (Standard)").
I don't know the origins, but we seem to use 'she' as a synonym for 'the situation'.

She'll be right. She's all good. She's apples.

etc.

Idiom: “she’ll be right, mate.” It’s generic, meaning “it will be alright.”

Despite all this, it still reckoned me en-NZ before en-AU. Oh, the shame! 😉

Being Australian, I selected that answer but it actually guessed I was a New Zealander first (2 and 3 were Australian and South African respectivly)
Yeah, that one was rather fishy.

It guessed correctly that I was Australian.

The other option was to ask "_________ Oi! Oi! Oi!"
I'm impressed - it suggested New Zealand as #1, and Australia as #2. Raised in Australia and New Zealand, with kiwi parents.
I got Australia #1, New Zealand #2. Born and raised in Australia and never been to NZ. Pretty similar dialects though I guess :)
It said I was from New Zealand. I've never been to New Zealand in my life.
I've noticed that prepositions seem to change usage over time in different regions. For example, "different to" has become common in British vernacular, even though "different from" is still typical for formal British English and in American vernacular. Sometimes I hear "different than" but I'm not sure where it is commonly used.
"It was the monkey that pushed the bear." What possible room for interpretation is there here? Unless you're not paying attention.

http://i.imgur.com/cxGR1Mm.png

I think this thing is broken. I did catch myself almost selecting something ridiculous-sounding a few times because I didn't read the sentence close enough, though; maybe that's what happened. But there were a ton of "US Black Vernacular" options that I never picked.

I guess most people from "ebonics" postcodes who took this quiz and trained it are more educated than their neighbors so they ended up picking the US standard English options like you did.
Wow, this is pretty good. It guessed my Finnish native language. Although it said my English dialect is 1. US Black Vernacular / Ebonics 2. Singaporean 3. New Zealand. Well, I don't think I have any specific dialect so it might be just pretty random.
totally incorrect for me. Didn't guess my nationality and put Russian at third place in native languages.
Other: impossible to complete on iOS due to insane drop down boxes that don't allow one to make or even view the proper selection. So, I suppose, I'm in the entitled American group. Finally, popular.

edit: dead I guess? If anyone has showdead on and sees this, I'd be much obliged if you could take a gander at my history and theorize as to what got me banned. Email in profile. Thanks in advance.

What it's testing for:

1. Passive alternation

2. It-clefting (simply `clefting` in some circles)

3. Another it-cleft?

4. It-cleft + possible scope ambiguity

5. Quantifier scope ambiguity (do you get quant raising, basically)

6. More clefts? Don't recall, but maybe they're mixing it up with the overt complementizer (`that`).

7. Ditto above

8. Passive construction again

9. More quantifier scope stuff

10. Lexicon inventory (modal shall is antiquated in most (all?) dialects)

11. Idunno, phrasing. Not-quite-collocate decisions

12. Aspect-tense interaction, I guess...

13. 'Conjugation', or how do you express tense?

14. Lexical selection/wh-feature spellout (does `+person` override the 'incorrect' case-marked `whom`?)

15. Skipped this one due to a double-click, too much coffee

16. Even more lexicon stuff

17. Lexical selection, or 'what kind of thing needs a determiner?'

...okay this is tedious. My point was going to be that a lot of these, especially the syntactic construction ones, seem very specific. I don't know of any dialects of English for example that switch the assignment of roles (A speech delivered Bill, eg) so freely as some of these questions suggest, and I thought all dialects had passive construction and it-clefting. Maybe I'll go through again and answer all the wrong things and see where it thinks those features are from.

Anyway, this was interesting. If this sort of thing also interests you, waste the rest of your day on The World Atlas of Language Structures! [1] It's a ton of fun.

[1] http://wals.info/

> I don't know of any dialects of English for example that switch the assignment of roles (A speech delivered Bill, eg)

This is a complete guess, but perhaps this exists to identify english-as-a-second-language speakers from languages where this form exists?

(English is my first and only language and I know very little about languages. The test guessed correctly that I am Australian)

Yeah, that's entirely possible.
Keep in mind that with surveys like this, especially online ones, it is important to have trap questions that will indicate if the responder is answering randomly and/or jokingly. So you may be correct that some answers are just wrong for everyone.
None of the top three guesses are correct in my case, not even close. I wonder if this is because they have more data for Chinese and Portuguese (Brazil) speakers.

    Our top three guesses for your English dialect:
    1. South African
    2. Welsh (UK)
    3. New Zealand

    Our top three guesses for your native (first) language:
    1. Chinese
    2. Portuguese
    3. English
Brazilian here and I got

1. US Black Vernacular / Ebonics

2. Singaporean

3. American English

Native:

1. Dutch.

2. American

3. Greek

They appear to list three guesses for variety of English for everybody who takes the quiz. They correctly spotted that I'm a native speaker of American (standard) English and listed the other two possibilities as Canadian (I do listen to radio from Canada, by rebroadcast on Minnesota Public Radio) and ebonics (which everyone in the United States hears sooner or later).

The guesses for my language background were spot on too. Like many Minnesotans, I have a few dialectal usages that reflect Scandinavian influence, and besides English as a first language, the other two languages guessed for me were Norwegian and Swedish. I have Norwegian ancestry (in part), so I come by those dialect habits honestly.

Counterdata: I've lived in California my entire life, my only other language exposure in day-to-day life is Spanish, and I have no Norwegian ancestry or local connections. It still thought my second- and third-most likely native languages were Swedish and Norwegian. I'm guessing those languages have ease picking up "native" English.
32 years in California, then 22 in Washington. I too got Norwegian and Swedish as the guess 2nd and 3rd guess for native language.

For my English variant, Standard American was first, then Canadian, then Singaporean.

Another Californian/Norwegian match here.

I also got Ebonics/Southern Black English as a second guess. I imagine that could be due to having family from Louisiana.

Hm. It guessed Finnish, Greek and Russian for my native language (German in reality) and a South African/Singaporean/New Zealand accent, where I have no idea how that could have happened – certainly not because I spent any amount of time in these countries (I didn’t) or consumed primarily content originating in these countries (also not the case, maybe apart from LotR…).

Finnish, Greek and Russian also don’t appear to have any connection to other Germanic languages…

Worked very well for me, correctly guessing my dialect which is spoken by relatively few people (not one of the 'standard' ones). I definitely found this surprising, especially considering I haven't lived in my home country for the better part of a decade. Very cool.
I'm from Australia. It thought I was Welsh. Apart from a strong sheep segment in the economy and being considered second-class by the English (and who isn't?), there's not much in common between the two countries... :)
Possibly because Welsh English is quite close to Australian English according to this blog post:

http://gameswithwords.fieldofscience.com/2014/05/the-english...

Interesting visualisation, but if I'm reading it correctly, a smaller number means more similar, and the number for Australian->Welsh is the third highest number for the Australian dialect, suggesting that they're more dissimilar than others.

Hrm, that's if you mouseover Australian. If you mouseover Welsh, Welsh->Australian is indeed the lowest number of the list. So, according to these numbers, Welsh is most like Australian, but Australian is least(ish) close to Welsh.

Weird.

Perhaps the Welsh emigrated here more prevalently than anywhere else (excluding the Welsh-speaking enclave of Patagonia!)

Anyhow, its guesses:

Nationalities: 1 Aus 2 NZ 3 Welsh Native Language: 1 English 2 Dutch 3 Swedish

I'm not sure where the germanic influence is supposed to have come from. The only foreign languages I've studied have been Latin, French and Spanish.

For the US, also check out the short quiz from the NY Times [1] based on the Harvard Dialect Survey from ~10 years ago.

It got both mine and my spouse's hometown area dead right.

[1] http://www.nytimes.com/interactive/2013/12/20/sunday-review/...

That was very interesting for me, an Australian living in the US. Phoenix specifically. I'm constantly amazed at how quickly American's can spot that I'm an Aussie.

I've gotten pretty used to the obvious different words between Australia and the US; rubbish = trash, car boot = trunk, pram = stroller, take away = carry out / to go, etc.

However this quiz highlighted a bunch that I wouldn't have been aware of, such as no one in the US knowing what soft-drink is and nature strip wasn't even on the list for "the grass beside a road".

One question in particular got me thinking that this quiz could also identify people with a scientific background, who read a lot of sci-fi and fantasy, or in general people not inclined towards future shock:

> Fill in the blank. Check all correct answers.

> The sun is in ________.

> ☑ the sky

> ☑ a sky

> ...

It wasn't until after I had finished that I realized I was mixing up two distinct concepts in my responses:

1) What I would consider correct phrasing for my own speech, and

2) What I would accept as correct when spoken by someone else