Copying the errors and making them even worse? Good job, Android!


A while back, there was a stir over Siri, the voice-activated search engine on iPhones, because it overlooked things like the addresses of Planned Parenthood clinics…which was a legitimate problem with the search engine behind it. So now there is a competitor on Android (oh, man, it is so pathetic that they named it “Iris”, Siri backwards) that is even worse. Take a look at its answers to some questions: it’s using a fundagelical search engine!

Comments

  1. onkundigna says

    Nitpick,
    Iris is an app built by Dexetra. You can hardly blame Android for the stupidity of app developers.

    That said, I do agree that the search engine is returning very stupid answers.

  2. es0tericcha0s says

    Unlike the iOS, Android doesn’t restrict the developers of it’s applications that much, so almost anyone can make an app to put in the Market. This is not an “Official” Android app that would come with a phone, but only something someone would have to choose to download from the Android Market. Sure, it’s stupid, but not Android’s fault. ;)

  3. Pierce R. Butler says

    I was going to write

    Turing Test: fail.

    but then I realized that there actually are humans that dumb.

  4. says

    Hm, where is Satan now? In hell, perhaps? Then where’s hell (and under the earth is occupied by rather dense material)?

    I would like to try out the follow-up questions, as they must become obvious fantasies, or circular, in a very short time.

    Glen Davidson

  5. michaeld says

    ‘Tis Himself, OM ,

    I our defense Harper got in with only 40% of the votes. So 3 out of 5 people didn’t want the ass.

  6. Zinc Avenger says

    It looks like it uses the “answer engine” ChaCha, which uses humans to give answers, which are then re-used when similar questions come up. So there’s a godbot filling the database with misinformation. Whoo!

  7. Drolfe says

    Right, ZA. But according to the article ChaCha claims to have many levels of editorial checking that are also powered by these sort of micro-wages that are also being basically gamed. Either hordes of godbots are approving questions at all levels, or their editorial process doesn’t work or is non-existent (editors just approve everything that crosses their path, i.e., automated click farmers or who knows).

  8. Rip Steakface says

    I was just about to say that ChaCha, which powers this app, uses humans to answer questions, often by doing a simple internet search, or in the case of a couple of these answers, by godbotting.

    This is pretty much exactly like Siri’s problem – this is a fault in the search engine, not because of Android, Google, or even the developer of the app. The Android Market, by the way, isn’t a closed garden like the Apple App Store. Android is open source (so far as I remember) and so anyone can develop for it and put whatever they want on the market. Sturgeon’s Law is in full force, but remember the positive side of it – the 10% of it that’s gold truly is gold.

  9. Alverant says

    What’s wrong with the first answer? We didn’t evolve from apes, we evolved from an ape-like creature. And technically humans ARE apes, but I’m not sure where our common ancestor fits in the current taxonomy.

    The other answers are still crap though, especially the last one. The middle two I can ignore because it’s a religious opinion (though I bet there would be an outcry if it gave anything but a christian answer) but saying it’s possible to justify rape, that’s just fucked up.

  10. PFC Ogvorbis (Yes, they are) says

    You can hardly blame Android for the stupidity of app developers.

    We most certainly can. Some testing using various ‘controversial’ phrases or questions from all sides of the polical and social landscape could very quickly expose biases which could then be corrected. Sounds like Android’s masters screwed up.

    If this is an actual Android product, major league fail. If it is an independent ap, they still have options.

  11. Drolfe says

    PFC Ogvorbis,

    I think to be fair in a case like this you can’t “blame Android” (the platform or its “App Market”). This is the trade-off you’re making when you choose between a walled-garden or a more open-market product. The open-market is open exactly because it’s stepping back and giving away the responsibility of taking care of ‘the garden’. Android is saying, hey, we’ll sell any old shit and buyers will float the cream to the top, first movers and early adopters will suffer but that’s “the price of freedom” (Wolverines!). When you turn around and pick the walled-garden instead, you’re choosing to have an arbiter, a gardener, who’s going to try and make sure the garden isn’t full of shit. The gardener is also a convenient target for blame. (“How could Apple let this shitty app through?!”) In this latter case blaming the vendor is fair.

  12. bartmitchell says

    PZ’s inner Apple Fanboi is showing again. This has nothing to do with Android, its hardware, or OS. As so many others have pointed out, it’s an independent app, produced by an individual. Android is about freedom. Apple is more like Syria, people can protest, but when it starts to cut into the power of the masters, they will use any force available to shut them down.

  13. Drolfe says

    I’m not sure what protest has to do with the analogy you’re trying to create, bartmitchell.

    I think I pretty clearly described the advantages and disadvantages of the two ploys. Openness means more shit, more shit like this, and to the proponents of openness that’s a feature, not a bug. Walled-garden means less shit, and someone to blame when shit is found, and someone to do something about the shit. It also means that you have to trust the gardener to not be evil. It’s OK to not offer that trust, there’s no judgement here.

  14. Rip Steakface says

    Openness means any jackass can put up a program for download or sale. It also means there’s no barrier to entry, so potentially good ideas don’t have to pass an approval process with potentially arbitrary requirements. A closed garden means a lot of the shit gets filtered out, but also means that the arbitrator can decide to not let whatever they don’t like into the market – and that you have to trust them to not be douches.

    I prefer openness if only because it allows for free programs that are easy to access and customize. A lot of the fun comes from customizing your device or app to your heart’s content.

  15. AlanMac says

    Where did homo sapiens originate?

    The first homo sapiens are thought to have originated in what is today Ethiopia. This was thought to be 3 mil. years ago. ChaCha!

    Where did humans originate?

    Where you believe that humanity originated depends on your belief about origins. If you believe in Creation, then you believe that humanity began in the Garden of Eden about 6,000 years ago.

    Nice try , but both answers are wrong.

    Where did the human species originate?

    Anthropologists calculate that modern humans, homo sapien, evolved 200,000 years ago in Africa.

    What a useless piece-o-crap

  16. Drolfe says

    Right, Rip Steakface made a crucial point I forgot to add in my section on openness, no gardener to get in the way of good ideas (for any reason)!

  17. n00blet says

    This is not Android’s answer to Siri, it’s merely an app that someone has developed and made available on the Android Market. Yes it has a stupid name, but in its defense it was apparently written as almost a joke, at most a proof of concept, and has taken off since then. It is not integrated into Android in any way, not core functionality.

    I hope this post’s subject is due to lack of knowledge/info (fixable) and not childish fanboyism – PZ is surely above that.

  18. FlickingYourSwitch says

    Seems unfair to blame the operating system itself or its developers, when iris is just a program that runs on that platform. There are no walled gardens in the world of Android, which is why there is more freedom there than on other platforms.

  19. Nemo says

    To be honest, if you’re turning to your phone’s AI bot for answers to these questions, you may have bigger problems.

  20. Amphiox says

    but then I realized that there actually are humans that dumb.

    It’s the new strategy for hard AI, you see. Making computers as intelligent, generally, as humans, has proven too hard.

    So enterprising researchers have decided to approach the problem from the other direction. Make the humans as dumb, generally, as the machines….

  21. unclefrogy says

    As I read the post I don’t think the comment is aimed at Android as much as it is aimed at stupid software giving that is giving out stupid religious answers to questions.
    I don’t have one of either of the popular handheld computer phones (cellphones)so I have a question.
    Do either of the search engines give similarly accurate answers to other kinds of questions?

    uncle frogy

  22. Amphiox says

    Harper grates on me, but frankly, if he were an American republican politician, he would have been primaried by the Tea Party and drummed out of politics long ago for being too moderate.

  23. ibyea says

    Is it just me, or are people these days getting more fond in voting for psychopathic conservatives?

  24. Kazim says

    For the record, I downloaded Iris and the first thing I said was “Open the pod bay doors Hal.”

    Gratifyingly it responded, “I’m sorry Dave, I can’t do that right now.”

    I’m willing to cut Iris some slack for a bit.

  25. Heliantus says

    What riles me up the most is actually the last question in the screenshot. Maybe because I’m just back from the Grammy thread.

    Q: “Is rape ever justified?”

    A: “Oh, it’s just a matter of opinion.”

    WTF? On top of everything else, the application is into relative morality? Or is it also from the same fundy website?

  26. municipalis says

    Amphiox:

    Harper grates on me, but frankly, if he were an American republican politician, he would have been primaried by the Tea Party and drummed out of politics long ago for being too moderate.

    I detest Harper, but if he were running in the U.S. he could easily fit in as a southern Democrat.

  27. nemothederv says

    So it’s using “family feud” technology.
    Y’know, using an average of most popular answers as opposed to factual ones.

  28. nemothederv says

    @12 Alverant

    I think it’s a case of stopped clock syndrome.
    Either that or someone went off message.

  29. aekendall says

    There’s a critical piece of information that I have not seen mentioned yet, but should change you way you look at these Isis answers.

    The way that “Answer” applications like Isis and Siri work is by using online search engines to find similar questions that have already been posted online. When the app finds some text that seems like a good fit for whatever was typed or said, it gives you that posted answer, whatever it may be. In other words, the answers that the app gives you are not a reflection of the developer or operating system. They are plucked from Joe Shmoe’s blog, or Yahoo Answers, or Wikipedia.

    Nor are the answers explained by the app exclusively using a “fundagelical search engine”. When I go to Google (hardly a Christian-biased search engine) and search for “is satan real” I get similar phrases in the first page of results.

    In case you think that the app developer built this (very specific) bias into their algorithm that fetches answers from search engines — despite the extra time and cost this would have required on their part with no direct financial benefit — I recommend this SlashGear article in which the tech reviewer finds many problems with Isis and speculates that it being created in 8 hours has something to do with it.
    http://www.slashgear.com/iris-voice-control-app-is-devs-8hr-attempt-at-siri-for-android-21189888/

  30. Paul W., OM says

    Alverant @ 12:

    What’s wrong with the first answer? We didn’t evolve from apes, we evolved from an ape-like creature. And technically humans ARE apes, but I’m not sure where our common ancestor fits in the current taxonomy.

    We didn’t just evolve from an ape-like creature—we evolved from the very same latest common ancestor as all the other apes, i.e., the one that it makes you an ape to be descended from. That species is now extinct, but was your basic model prototype ape in any reasonable scientific sense.

    We aren’t just “technically” apes, we are apes in the only sense in which “ape” is a clearly scientific category.

    More interestingly, we are also monkeys in the very same basic and important sense, which is the only clearly scientific sense.

    We are more closely related to Old World monkeys than Old World monkeys are to New World monkeys. We apes evolved from the very same latest common ancestor as the (other) monkeys, and we are all more closely related to each other than to some other primates, like tarsiers.

    It’s basically false to say both that

    (1) we are apes (or evolved from apes), and

    (2) we apes aren’t monkeys (or didn’t evolve from monkeys).

    Anybody who already understands the obscure and very different technical senses in which those things are both true doesn’t need to be told.

    Anybody who doesn’t already understand these things will be systematically misinformed.

    Better to admit that, yes, of course, an ape is a kind of monkey, as people have always tended to assume just from looking at them.

    Unfortunately, a lot of science educators and popularizers think it’s important to “correct” the “mistaken” idea that apes “are monkeys,” without ever getting across the obscure and gerrymandered sense in which they’re not—i.e., that apes are the monkeys we don’t call “monkeys”, because that’s the rule, that’s why, even if if the pesky science says you are too a monkey.

  31. says

    You would see the same exact responses in an iOS app too. Apple doesn’t censor stupidity. They refuse programs that contain nudity, racism, malware, privacy invaders, and duplicate existing iOS features (music, dialars, launchers etc). Last year they banned an app that was created as criticism of the smart phone development process. The gaim contained child labor, worker suicides etc.

  32. christophburschka says

    It started off okay with “humans are more closely related to apes than to monkeys, but we didn’t evolve from apes either” (which is the correct answer when talking about “modern apes”, so it might be worth the benefit of the doubt) then went a bit too wishy-washy with the “only you can answer”, but I didn’t understand what was so bad until—

    “… satan is real …” what the—

    “… rape … a matter of opinion …” FUUUUUU-

    Note that ChaCha is a human-powered engine. Its user base, not the Android platform, is to blame for the answers themselves, but the developers should definitely have added “not populated by idiots” to their criteria for choosing the web service. Maybe Yahoo Answers might have been a better choice:

    “how is babby formed. how girl get pragnent”

    “they need to do way instain mother>”

  33. says

    So, I read PZ’s post and wanted to say that “Good job, Android!” was inaccurate for because this is an independent app by a random Indian company called Dexetra and Android is not a closed garden so that it is available doesn’t mean an endorsement for it.

    …Then I noticed that it has been already addressed in comments. So I decided not to post it again and insist on the discussion.

    …Then I saw people insisting on this topic and for some reason taking jabs at PZ for it, please guys, come on… Chances are anyone actually reading the comments already knows more than enough about the issue, you may focus on other things now.

    Like for example, what about someone makes an App equivalent that does not do these stupid things? And why is Siri still endorsed by apple?

  34. says

    @37: aekendall , I guess this demonstrates that the internet is stupid.

    These answer engines are unlikely to ever work correctly. A tempting solution would be to ban sites that are obviously going to give stupid answers, like all creationist and anti-vax sites. But then what to do with answers from legit sites that are still wrong? Validating them is going to be quite a nightmare.

    Of course, they could make the app show up all possible results. Perhaps indixed and let them click the links of the origins of the sites to find out what was going on. They can also rename the app to “web search engine”.

    I also wonder if these apps are even legal. They just extract info (and traffic) from other sites…

  35. says

    I just tried this, and here’s my result:

    Q: Is global warming real?

    A: No, Global Warming was created by politicians in Washington. They paid scientists to invent the idea of global warming.

  36. Paul W., OM says

    Oh, and FWIW, it’s common in most languages to use the word for “monkey” inclusively, to include apes. E.g., in Dutch or Swahili or Latin, an ape literally is a kind of monkey, both vernacularly and scientifically.

    That was a common, accepted usage in English as well, even among scientists studying such animals, up until the 1920’s. It was not “wrong,” just a manifestation of the usual sort of ambiguities you get in category terms.

    E.g., if you’re attacked by an ape, you’d probably say so, rather than saying you were attacked by “a monkey.” The latter would not have been incorrect, just confusing, because calling it a “monkey” suggests that it was probably not an ape.

    Likewise, even now it would be weird to say that you were attacked by “an ape” if you were attacked by an ape you recognized as a human. Not technically false, given that humans are apes, but misleading because it’s bizarrely uniformative.

    IMO we ought to go back to the old way. It was just stupid to try to define “monkey” as exclusive of “ape,” and pound into people’s heads that “chimps aren’t monkeys.”

    Notice that these days we do tell people that “humans are apes,” picking a side in a vernacular language ambiguity.

    Of course we are not apes in the obvious vernacular sense that “apes” are by definition not human—they’re things like gorillas or chimps but not humans.

    Going ahead and telling people that yes, humans are apes is thus informative—of course we’re not “apes” in that exclusive, superficial vernacular sense, but we are nonetheless apes in a scientifically basic and important sense you should know about.

    It wouldn’t be useful to tell people that humans are not apes, because they already know that they’re not “apes” in that vernacular sense—they’re not that hairy, they stand upright, etc. Telling them that they are nonetheless apes is informative.

    Similarly, it’s not very informative or useful to tell people that apes “aren’t monkeys.” (Which is different from saying that most monkeys aren’t apes.) Without explaining the basic sense in which they are monkeys first, it can only confuse them.

    christophburschka:

    It started off okay with “humans are more closely related to apes than to monkeys, but we didn’t evolve from apes either” (which is the correct answer when talking about “modern apes”, so it might be worth the benefit of the doubt)

    I very much disagree. If we’re going to tell people about any of this stuff, and pretend to get it right, we should actually get it at least basically right in terms of the basic scientific facts.

    No, we didn’t evolve from extant nonhuman apes, but we certainly did evolve from apes. Apes that you would immediately recognize as apes if you saw one. (Likewise those apes evolved from monkeys that you’d immediately recognize as monkeys.)

    We simply did, and saying we didn’t is utterly false. It makes it sound like apes and humans don’t have a common ancestor that was recognizably an ape, but they do. It might not have been exactly a chimp, and it wasn’t a gorilla or a siamang, but it was very definitely an ape.

    If people don’t understand the idea of an extinct ape, they need to be told.

    It’s ridiculous to restrict “ape” to mean “extant” ape, if we’re trying to explain categories like “ape” scientifically. (And if we’re not going to do that, we should shut up about it rather than spreading worse confusion.)

    Notice that we have no problem talking about extinct species of, say, birds or antelopes or horses.

    We would never say that finches didn’t evolve from other birds just because the “birds” they evolved from are all extinct. “Bird” and “extant bird” are just not the same idea.

    Saying that we “didn’t evolve from apes” is the same sort of thing. If extinct ape ancestors don’t count as apes, you’re missing the whole point of evolutionary category terms, and should give up and go home.

    When answering a question, it’s always good to think about what people are actually interested in.

    People are interested in whether humans evolved from apes. They are not just interested in whether they evolved specifically from chimps, but in the more general question of whether they evolved from some kind of ape.

    The answer to that is unambiguously yes, not no.

    You can say “yes but not any extant ape” if you want, but the answer has to be “yes,” not “no.”

    Similarly, people are not especially interested in the questions of whether humans evolved from macaques or baboons or spider monkeys, specifically. They don’t know the difference between platyrrhine and catarrhine monkeys, and wouldn’t care much if they did.

    They want to know if we evolved from some sort of monkey, and any early simian would do just fine.

    The answer to that has to be yes. We evolved from catarrhines that we would immediately recognize as monkeys if we ran into them in a zoo.

    If you realize that, and tell people that we “didn’t evolve from monkeys” it’s a lie.

    They’re not asking whether we evolved from extant monkeys. They mostly haven’t thought much about that, but if they did it wouldn’t change anything much. They’d still want to know if we evolved from monkeys, and answer has to be yes.

    We should answer the question that was asked, not some other much less interesting question that was clearly not the one asked about a less interesting category of things.

    The questions of whether we evolved from apes or evolved from monkeys are easy ones. There’s no reason to think that the nearest common ancestors were radically different from most of the extant descendant species in either case, and there are pretty good reasons to think they weren’t.

    It’s funny how willing we are to give the basically wrong answers when it comes to human origins. In similarly clear cases, we wouldn’t hesitate to call a dead horse a horse. (Of course.)

  37. says

    Paul W.,

    please don’t make a statement about “most languages” based on 3 languages. 3 out of 6000-8000 languages are not “most”.

    Also I don’t know how you can imply “an ape is a kind of monkey scientifically” in one language but not in another.

    Vernacular usage is not always clear:

    In Chinese and Dutch, a whale is a fish etymologically, but due to science education, most people would no longer hold that.

    Or Samoan uses the term “fish” for all being that live in the sea, and “bird” for anything that flies. This is actually taught as a Samoan way of looking at nature, but then alongside that a scientific view is taught in school as well.

  38. says

    Well, if you play with chacha for a while, you will see that it is a mixed-bag. There are humans answering the questions and what you get depends on who answers it. So I’d guess the bias would average out to the bias of the employees who answer the questions.

    Q: Are atheists immoral?

    A: Not at all. Many have a high standard of values, morals, and how to live. They just don’t believe in a divine creator. ChaCha!

    Q: Is capital punishment a sin?

    A: After the flood of Noah, God instituted capital punishment, commanding, “surely for your lifeblood I will demand a reckoning; …Whoever sheds man’s blood, by man his blood shall be shed; for in the image of God He made man” (Gen. 9:5-6).

  39. aekendall says

    @42: vexorian, Yes, the internet is very misinformed about many topics. Of this there is much evidence!

    You raise a good point: it is very, very hard to get these kinds of “Answer” search engines to work correctly — or at least, meet the expectations of a random competent internet user who can use Google.

    As technical tasks go, building a program in 2012 that performs a web search is relatively easy. The truly hard part comes from 1) knowing exactly what to search for, and 2) what to do with the many results you get.

    Obviously, Iris is failing at this critical part. It appears that it even has trouble finding answers to the most basic questions that have clear factual answers. What is 12 + 3? How many grams in an oz?

    Throw a complex political question at it?

    It doesn’t have a “prayer”.

  40. Paul W., OM says

    pelamun:

    please don’t make a statement about “most languages” based on 3 languages.

    I didn’t. I just gave 3 specific but varied particular examples—an African language, a Germanic language, and a romance language—to illustrate my point. There’s a bunch of varied others, or so I’ve read, (e.g., Hindi and German and at least some Slavic languages, IIRC…?)

    For my arguments, it’s not essential whether this is a cultural universal, or even typical, or just reasonably common in a reasonable variety of languages and cultures.

    It may not be the only natural way of dividing things up, but it’s clearly not an unnatural way that’s difficult to learn.

    The point is that it’s pretty natural to look at monkeys, and look at apes, and think of apes as a kind of monkey, whether or not you know the evolutionary relationships between them. And if you do know the evolutionary relationships, it doesn’t seem scientifically wrong; it seems quite right. In this case, appearances are revealing. (And having the language fit the naive appearances is thus a plus.)

    Another point is that the arbiters of “correct terminology” in quasi-scientific mid-20th-century English have chose to go one way, but could quite reasonably have gone another. There was plenty of precedent, cross-culturally, and specifically in English, and more specifically among English-speaking scientists, for using the term “monkey” inclusively of apes, to mean “simian.”

    It’s not a matter of scientists imposing a scientifically important distinction on vernacular language, as when saying that “whales are not fish.” The important scientific distinction clearly goes the other way—the actually important, actually scientific term “simian” is inclusive of apes.

    What’s weird here is that science educators and popularizers chose to try to standardize a vernacular not-really-scientific term in a confusing way that directly conflicts with the actual scientific term.

    They could have chosen to say that “monkey,” strictly speaking, is just the vernacular version of “simian.”

    Instead they chose, in the face of a vernacular ambiguity, to standardize the vernacular term in the other way, and obscures the scientific facts.

    I suspect that was largely for the convenience of scientists, who generally know both words and can say “simian” when they want an inclusive term, and “monkey” when they want to exclude apes. They were in the habit of calling both OWM’s and NWM’s (but not apes) “monkeys,” whether it made any real sense or not, and didn’t need to clean up their terminological act because they had “simian” to use as well, among themselves.

    So they exported their fucked-up, arcane use of the term “monkey,” standardizing it for non-scientists as well, and kept the better term, “simian,” for themselves.

    That’s bad for everyone else. Most people are never clear on the term “simian” and don’t ever use it, and are stuck with the screwed-up, gerrymandered version of “monkey” instead.

    (Even those of us who do know the word “simian” can’t usually use it, even among most scientists, because we can’t count on nonspecialists knowing it.)

    Scientists are wrong to tell people that apes are not monkeys as though it was a fact, rather than an artifact of their own fucked up version of casual, vernacular terminology—which only makes sense, even for them, because they have other, better words for the basic facts.

    Also I don’t know how you can imply “an ape is a kind of monkey scientifically” in one language but not in another.

    What I mean is that the common words corresponding roughly to “ape” and “monkey” actually correspond closely to “ape” and “simian.” They have a word for apes, and a word for monkeys-including-apes, rather than mutually exclusive terms for (nonape) monkeys vs. apes.

    (E.g. IIRC, in at least some Germanic languages, the general term is a cognate of “ape” but means simian, and the more specific term for what we call “apes” in English is something like “man-ape,” meaning manlike simian.)

    That terminology fits the biological categories, because there’s a natural subcategory relation there—apes are simians, but not all simians are apes, because apes evolved from simians, not vice versa.
    The terms are asymmetric because the basic facts are asymmetric.

    Saying that “apes are not monkeys” and that they evolved from some vague “common ancestor” obscures the crucial facts that

    (1) apes evolved from monkeys, and

    (2) there’s no really scientific sense in which they ever stopped being monkeys.

    Even most scientifically literate people don’t know these things, and I suspect that’s largely because they’re often falsely told the opposite: that apes somehow aren’t monkeys, and even that they evolved from some “common ancestor” that somehow was also “not a monkey.”

  41. Paul W., OM says

    pelamun:

    I’m not sure I get the significance of what you were saying about (e.g.) fish in (e.g.) Samoa.

    I’d be a whole lot less bugged by all this if I thought most people actually were taught the concept of “simian” (monkeys inclusive of apes), and the actual relationships between clades, as well as the vernacular factoids “chimps aren’t monkeys” and “we didn’t evolve from monkeys.”

    The sad fact is that in the U.S., at least, even most scientifically literate people clearly do not learn the relevant actual facts. They don’t know what a “simian” is, they don’t know how or why apes are officially excluded from the so-called “monkeys,” they don’t know that the common ancestor of apes and monkeys was a monkey, etc.

    They do learn that “chimps aren’t monkeys,” and “humans didn’t evolve from monkeys,” which tells them a whole lot less than nothing of importance. As they understand the terms, the former statement is pointless and the latter is just grotesquely false.

    This is even true of most scientists who aren’t biologists. Many pro-evolution scientists in places like Pharyngula will parrot factoids like “humans didn’t evolve from monkeys” as though they were straight scientific facts, as opposed to being straighforwardly false on any straightforward interpretation of terms.

    That sucks.

    Unless and until people start learning the actual facts despite the fucked up terminology, I’ll advocate reforming the terms to reflect the facts.

  42. says

    Paul W.,

    you haven’t even shown that it’s reasonably common, because all your examples but Kiswahili are from only one language family, i.e. Indo-European. Linguists have spent some time on developing sufficiently diverse language samples.

    But it also ignores the fact that in many languages, there are no hyperonyms available. I ran into this doing my research, that in a language I was working on there were no words for “birds”, or “animals”. Instead you’d need to enumerate the different species individually. I’d just suggest avoiding making sweeping statements about all languages/cultures, is all.

    Also, you always need to differentiate between the scientific terminology, and the everyday speech. Languages that have been used for academic and scientific purposes (the number is probably about 100-150) have developed terminologies for biology as well and will have scientific terms available for “ape”, “monkey”, “simian”, “primate”. But that doesn’t tell us anything about common popular usage, because all scientific languages will have the terms for the plethora of orders and subdivisions, which will be well beyond what you can achieve by even just differentiating two words, “monkey” and “ape” in everyday language. Why not use a third word, more common in English, like “primate” instead? I don’t know any language that would even have a common word for “simian”, they all seem to be complex compounds. (BTW, is “simian” and “prosimian” still used? Wikipedia keeps telling me in different languages that it is haplorhini (“dry-nosed apes” in German) and strepsirhini (“wet-nosed apes” in German) now?)

    Words have several interconnected meanings, while their meaning tend to be very precise in an academic context, this won’t be the case in everyday speech.

    Basically I see thee different approaches to this problem:

    1. there are no hyperonyms, each species (or perceived species) has to be named as such. Indonesian is probably a case for this, though there are two more general terms, monyet, roughly for “monkey”, and kera roughly for “ape”. I suspect that with the development of a scientific vocabulary (in the case of Indonesian, usually directly from the Latin) and globalisation (=influence of English language), there are attempts at mirroring monyet to English “monkey” and kera to English “ape”, but it’s my understanding that in popular usage, orang utans for instance would not be called kera.

    2. the language has one hyperonym, which is equivalent to “primates”. Japanese is a case of this, there is only one word natively, 猿 saru. This is also used scientifically for “primates”, though there is also the same word Chinese has coined for “primates” (see below). Distinctions are only drawn in scientific language.

    2a. the language has one hyperonym traditionally, but has developed a derived form which can be used for “apes”. German would be this. It uses Affe as hyperonym. Colloquially, Affe can be used for “primates”, though scientifically, “Affe” refers to “simians”. For “apes”, “Menschenaffe” was coined.

    3. the language has two different terms traditionally, one for “monkey”, one for “ape”. Chinese and English are examples for this. Chinese has 猴 hóu “monkey” and 猿 yuán “ape”. For the scientific terms, they appear as compound elements. 猴 hóu exclusively for prosimians, and both for simians (though for hominids 人 rén “human” is used). For “primates”, they created a new word: 靈長類 língzha3nglèi.

    It seems that if a language had two terms traditionally, language planners were reluctant in assigning one of them for all primates. Besides English and Chinese, Spanish (simio, mono) would be another example for this. I’d recommend using primates.

    Swedish and other Scandinavian languages as well as Dutch are like German, they have an ape-cognate (“apa”, “aap”) as the most general term and have coined a derived form for “ape” (“människoapa”, “mensaap”).

    French seems to be like Japanese, just that “singe” cannot refer to all primates, just to all “simians”.

    This sample is of course not representative, but I think it shows that not widening “monkey” to mean “simian” is not just some accident.

  43. says

    Paul W.,

    why I referenced Samoan: traditionally, they have classified “everything that swims” as “fish” (i’a) and “everything that flies” as “bird” (manu).

    Now instead of telling students that the traditional understanding of i’a and manu is “wrong”, they now teach both, making sure they understand the difference between the traditional i’a/manu and that scientifically, bats aren’t manu (AFAIK they use the same terms), and that whales and dolphins aren’t i’a.

    Similarly, instead of redefining “monkey” and “ape”, one should respect the traditional usage, i.e. “monkey : small” and “ape : big”, and teach the scientific categories alongside of it.

    Pushing the terms “simian” and “primate” might help, but apparently especially simian hasn’t gained much currency yet.

  44. gravityisjustatheory says

    pelamun says:

    why I referenced Samoan: traditionally, they have classified “everything that swims” as “fish” (i’a) and “everything that flies” as “bird” (manu).

    That’s not unique to Samoan. I’m pretty sure English used to classify things that way., and only changed relatively recently (probably in the last few hundred years).

  45. says

    gravityisjustatheory,

    are you just “pretty sure”, or do you know for certain. English happens to be the most researched language in the world, if you make such a statement about it, please provide references (For Samoan, you’ll have to take my word for it, sorry, but I studied it for some years back in university).