Linguistic pet peeves

A forum for discussing linguistics or just languages in general.
Post Reply
GrandPiano
mayan
mayan
Posts: 2080
Joined: 11 Jan 2015 23:22
Location: USA

Linguistic pet peeves

Post by GrandPiano »

What are some pet peeves you have about some people's ideas about languages and linguistics? It can be about one specific language or it can pertain to all of linguistics, and it can have to do with things that non-linguistic people tend to think or things that you find in published papers on linguistics (if you read that sort of thing).

In particular, I hate it when people think that voicing is more important in distinguishing English occlusives (by which I mean plosives and affricates) than aspiration just because the voiceless occlusives become unaspirated after /s/ and sometimes in unstressed syllables (should /æ/ and /ε/ not be considered distinct phonemes because they merge in unstressed syllables?). I never had any problem distinguishing between the Mandarin occlusives, which are distinguished only by aspiration, and before I got into linguistics I wasn't even sure whether there was a difference between the English and Mandarin occlusives! (I mean, I didn't think of them as "occlusives", but you get the idea)
User avatar
Ahzoh
mongolian
mongolian
Posts: 4191
Joined: 20 Oct 2013 02:57
Location: Canada

Re: Linguistic pet peeves

Post by Ahzoh »

That prescriptivist spelling standards means that languages can't eventually change.
Image Śād Warḫallun (Vrkhazhian) [ WIKI | CWS ]
clawgrip
MVP
MVP
Posts: 2257
Joined: 24 Jun 2012 07:33
Location: Tokyo

Re: Linguistic pet peeves

Post by clawgrip »

People say voicing is more important than aspiration because it is. If you say [pæk] most people will be unsure if you're saying "back" or "pack", but no one will be confused about what word you mean if you say if you say [spʰik].
Sumelic
greek
greek
Posts: 566
Joined: 18 Jun 2013 23:01

Re: Linguistic pet peeves

Post by Sumelic »

Actually, I have a bit of a peeve about that too [>:D] but in fact, I'm even more radical: I don't even like the concept that English has "allophony" between [pʰ tʰ kʰ] and [p˭ t˭ k˭] after /s/. My reason is, there is no process in English where you can take an aspirated consonant and then put an /s/ before it in the same syllable, causing it to become unaspirated. To me, allophony has to be proved through productive phonological processes in the language: for example, in American English, you can add a suffix to /kʰɹiˈeɪt/ and get [kʰɹiˈeɪɾəd], so that's good evidence that /ɾ/ is an actual, synchronic allophone of /t/ in American English.

But the tenuis stops exist solely as part of unanalyzable clusters; it seems ridiculous to me to say that they must somehow be underlyingly "the same sound" as the aspirated stops. After an /s/ there is actually a complete neutralization of the contrast between aspirates and the "voiced" (actually, often somewhat devoiced) consonants, so it is inappropriate in my opinion to consider this a definite allophone of either specific sound. The people who do so are just letting their knowledge of the phonological history, or more insidiously, the orthography, warp their perception.

In fact, there's a stronger argument for identifying them with the "voiced" stops:
a) many people (myself included) devoice voiced stops after /s/ in words like "disgust"
b) English already has common, productive progressive assimilation to voicelessness, as shown by the plural and past tense affixes
c) When dealing with other languages that have voiceless aspirates and voiceless tenuis, native English speakers tend to identify the voiceless tenuis with English voiced stops, not English voiceless aspirates, as you've experienced for yourself

Anyway, that's my big rant about this "allophony".

Another peeve I have is about two types of things people say about English spelling reform: the people who lie about the level of irregularity or capriciousness in the current system to exaggerate its flaws, like the famous "ghoti" example that could never actually be pronounced "fish", and the people who deny the obvious stupidities that do exist by claiming that all of the irregularities in our current orthography serve a deeper purpose, like etymology or distinguishing homophones. (You'll never be able to convince me that "aisle", "receipt", "friend", "aunt", "scythe" are optimal spellings.)
Last edited by Sumelic on 27 Mar 2015 03:31, edited 7 times in total.
Sumelic
greek
greek
Posts: 566
Joined: 18 Jun 2013 23:01

Re: Linguistic pet peeves

Post by Sumelic »

@clawgrip:
Your example cuts both ways. Since [pæk] is ambiguous, rather than clearly /pæk/, this shows that aspiration is just as important as voicing for identifying voiceless aspirated stops at the start of a word. In fact, English speakers are more likely to hear completely tenuis stops as voiced than as voiceless [for example, if you modify a sound clip of an English voiceless aspirated stop to remove the aspiration], because the so-called "voiced" stops of English are actually devoiced anyway for many speakers in word-initial or word-final position.

As for [spʰik], it's clear, but it's also something no native English speaker would ever say. As I outline in my post, in s-clusters there is total neutralization of the manner of articulation of the second element, so there's no way to get confused with any of [spʰik] [sp˭ik] or [sbik]. Even [zbik] would probably be recognized as "speak", since there aren't any similar consonant clusters it could contrast with.

Would you say that is just an allophone of /z/ in the initial clusters [sp-] [st-] [sk-] ?
Last edited by Sumelic on 27 Mar 2015 04:25, edited 2 times in total.
GrandPiano
mayan
mayan
Posts: 2080
Joined: 11 Jan 2015 23:22
Location: USA

Re: Linguistic pet peeves

Post by GrandPiano »

If you asked me to identify [pæk] as either <pack> or <back>, I would almost certainly hear it as <back>. In fact, another argument I forgot to mention for aspiration being more important than voicing is that I often find myself devoicing word-initial voiced stops and affricates.

I think some dialects actually pronounce the /t/ in <sixteen> and <fifteen> as unaspirated, but my dialect doesn't do that.

As for Sumelic's thing about spelling reforms, an idea I've had for an English spelling reform for a while is that it wouldn't be a complete spelling reform, just removing unnecessary letters and making certain rational changes, as well as leaving out certain changes in some words in order to not create homographs. Some ideas:
  • Distinctions that are still maintained in some dialects such as <w> vs. <wh> and <er> vs. <ir> vs. <ur> will be kept.
  • Double letters that are necessary to determine the pronunciation of certain vowels are kept doubled, e.g. the <dd> in <ladder> or the <mm> in <hammer>.
  • Silent letters are removed, so that <debt> becomes <det> and <island> becomes <iland>.
  • Letters can be added or changed when the original spelling makes no sense. For example: <move> becomes <moov> (not <muve>, because some speakers would pronounce that /mjuv/), <to> becomes <tu>, and <comb> would become either <come> or <coam> (I'm not sure what to do with <come>... The only good spelling for that I can think of has certain connotations...). <two> will probably keep its spelling to avoid homography.
  • If a double letter is unnecessary, it will be removed - <stack> becomes <stac>, but <stacking> keeps its original spelling.
  • You know the annoying irregularities of the "i before e, except after c" rule? It's all <ie> now.
  • "I" is no longer capitalized. Really, it's not necessary.
That's a bit of it. As you can see, there's still ambiguity, but it significantly improves the spelling without making things too weird (I suppose words ending in "oov" looks pretty weird, but I think we can get over that) or creating additional homographs. In fact, most words won't be changed at all.
Spoiler:
Or, in the reform:

That's a bit of it. As you can see, there's stil ambiguity, but it signifficantly improovs the spelling without making things too wierd (I supose wurds ending in "oov" looks pritty wierd, but i think we can get over that) or creating aditional homographs. In fact, moast wurds woan't be changed at al.
User avatar
Ahzoh
mongolian
mongolian
Posts: 4191
Joined: 20 Oct 2013 02:57
Location: Canada

Re: Linguistic pet peeves

Post by Ahzoh »

GrandPiano wrote:If you asked me to identify [pæk] as either <pack> or <back>, I would almost certainly hear it as <back>. In fact, another argument I forgot to mention for aspiration being more important than voicing is that I often find myself devoicing word-initial voiced stops and affricates.

I think some dialects actually pronounce the /t/ in <sixteen> and <fifteen> as unaspirated, but my dialect doesn't do that.

As for Sumelic's thing about spelling reforms, an idea I've had for an English spelling reform for a while is that it wouldn't be a complete spelling reform, just removing unnecessary letters and making certain rational changes, as well as leaving out certain changes in some words in order to not create homographs. Some ideas:
  • Distinctions that are still maintained in some dialects such as <w> vs. <wh> and <er> vs. <ir> vs. <ur> will be kept.
  • Double letters that are necessary to determine the pronunciation of certain vowels are kept doubled, e.g. the <dd> in <ladder> or the <mm> in <hammer>.
  • Silent letters are removed, so that <debt> becomes <det> and <island> becomes <iland>.
  • Letters can be added or changed when the original spelling makes no sense. For example: <move> becomes <moov> (not <muve>, because some speakers would pronounce that /mjuv/), <to> becomes <tu>, and <comb> would become either <come> or <coam> (I'm not sure what to do with <come>... The only good spelling for that I can think of has certain connotations...). <two> will probably keep its spelling to avoid homography.
  • If a double letter is unnecessary, it will be removed - <stack> becomes <stac>, but <stacking> keeps its original spelling.
  • You know the annoying irregularities of the "i before e, except after c" rule? It's all <ie> now.
  • "I" is no longer capitalized. Really, it's not necessary.
That's a bit of it. As you can see, there's still ambiguity, but it significantly improves the spelling without making things too weird (I suppose words ending in "oov" looks pretty weird, but I think we can get over that) or creating additional homographs. In fact, most words won't be changed at all.
Spoiler:
Or, in the reform:

That's a bit of it. As you can see, there's stil ambiguity, but it signifficantly improovs the spelling without making things too wierd (I supose wurds ending in "oov" looks pritty wierd, but i think we can get over that) or creating aditional homographs. In fact, moast wurds woan't be changed at al.
I pronounce "comb" with a "b" at the end.
Image Śād Warḫallun (Vrkhazhian) [ WIKI | CWS ]
GrandPiano
mayan
mayan
Posts: 2080
Joined: 11 Jan 2015 23:22
Location: USA

Re: Linguistic pet peeves

Post by GrandPiano »

Ahzoh wrote:
GrandPiano wrote:If you asked me to identify [pæk] as either <pack> or <back>, I would almost certainly hear it as <back>. In fact, another argument I forgot to mention for aspiration being more important than voicing is that I often find myself devoicing word-initial voiced stops and affricates.

I think some dialects actually pronounce the /t/ in <sixteen> and <fifteen> as unaspirated, but my dialect doesn't do that.

As for Sumelic's thing about spelling reforms, an idea I've had for an English spelling reform for a while is that it wouldn't be a complete spelling reform, just removing unnecessary letters and making certain rational changes, as well as leaving out certain changes in some words in order to not create homographs. Some ideas:
  • Distinctions that are still maintained in some dialects such as <w> vs. <wh> and <er> vs. <ir> vs. <ur> will be kept.
  • Double letters that are necessary to determine the pronunciation of certain vowels are kept doubled, e.g. the <dd> in <ladder> or the <mm> in <hammer>.
  • Silent letters are removed, so that <debt> becomes <det> and <island> becomes <iland>.
  • Letters can be added or changed when the original spelling makes no sense. For example: <move> becomes <moov> (not <muve>, because some speakers would pronounce that /mjuv/), <to> becomes <tu>, and <comb> would become either <come> or <coam> (I'm not sure what to do with <come>... The only good spelling for that I can think of has certain connotations...). <two> will probably keep its spelling to avoid homography.
  • If a double letter is unnecessary, it will be removed - <stack> becomes <stac>, but <stacking> keeps its original spelling.
  • You know the annoying irregularities of the "i before e, except after c" rule? It's all <ie> now.
  • "I" is no longer capitalized. Really, it's not necessary.
That's a bit of it. As you can see, there's still ambiguity, but it significantly improves the spelling without making things too weird (I suppose words ending in "oov" looks pretty weird, but I think we can get over that) or creating additional homographs. In fact, most words won't be changed at all.
Spoiler:
Or, in the reform:

That's a bit of it. As you can see, there's stil ambiguity, but it signifficantly improovs the spelling without making things too wierd (I supose wurds ending in "oov" looks pritty wierd, but i think we can get over that) or creating aditional homographs. In fact, moast wurds woan't be changed at al.
I pronounce "comb" with a "b" at the end.
Oh, seriously? I thought that all dialects had dropped that. See, this is why trying to reform English spelling while taking all dialects into account is hard. [:P]

My other plan is to just wait until English has split into American and British before planning any reforms.
Sumelic
greek
greek
Posts: 566
Joined: 18 Jun 2013 23:01

Re: Linguistic pet peeves

Post by Sumelic »

I’ve actually been researching the etymologies of English words for a little while to try to come at spelling reform another way, seeing if spelling words more in line with their etymologies would also improve their pronounceability. Surprisingly, fairly often it does! In my view, English spelling has only gone downhill since the days of Chaucer; his orthography was more sensible than ours.

And if you use the etymology of words as your basis, nobody can accuse you of being biased towards any particular modern dialect. :mrgreen:

The same text according to my reform:
Spoiler:
Igh’ve actually been recerching the etymologies of English words for a litel whyl to trie to cume at spelling reforme another wey, seeing if spelling words moar in line with their etymologies would also improove their pronounceabilitë. Surprisingly, fairly often it does! In my vew, English spelling has oanly gon dounhill sins the days of Chaucer; his orthographie was moar sensibil than ours.

And if ew use the etymologie of words as eur basis, nobody can accuse ew of being biased towards any particular modern dialect.
Eur text spelled according to my reforme: (purely etymologically based):
Spoiler:
GrandPiano: That's a bit of it. As ew can see, thear's still ambiguïtë, but it significantly improoves the spelling without making things too wierd (igh suppose words ending in "oov" looks pretty wierd, but igh think we can get over that) or creating additional homographs. In fact, moast words woun’t be changed at all.
GrandPiano
mayan
mayan
Posts: 2080
Joined: 11 Jan 2015 23:22
Location: USA

Re: Linguistic pet peeves

Post by GrandPiano »

That's pretty cool. Besides dialect bias, another big worry I always have in trying to design a new orthograpy for English is that the new spellings will look a lot like the kind of misspellings you see in our current orthography, for example "moov" instead of "move" or "wurd" instead of "word". In a morphophonemic spelling reform I've designed for my dialect, the <-s> suffix becomes <-z>, because that's how it's realized after vowels. However, replacing <s> with <z> is a pretty classic way to misspell words in our current system.
Spoiler:
In the other reform that I mentioned above:

Dhat'z pritī kūl. Besáydz dáyelēkt bayes, enedher big werī ay ālweyz hav in traying tū dezáyn e nū orthàgrefī for Inglish iz dhat dhe nū spēlingz wil luk e lāt layk dhe kaynd ev misspēlingz yū sī in awr kerent orthàgrefī, for ēgzampel "moov" instèd ev "move" or "wurd" instèd ev "word". In e mórfōfenìmik spēling rīform ay'v dezáynd for may dáyelēkt, dhe <-s> sefiks bīkémz <-z>, bīkóz dhat'z haw it'z rìelayzd after vawelz. Hawēver, repleysing <s> with <z> iz e pritī klasik wey tū misspèl werdz in awr kerent sistem.
GrandPiano
mayan
mayan
Posts: 2080
Joined: 11 Jan 2015 23:22
Location: USA

Re: Linguistic pet peeves

Post by GrandPiano »

Oh, and another big pet peeve of mine: When someone gives a list of written words and/or phrases in a language and doesn't say anything about how to pronounce them.
Sumelic
greek
greek
Posts: 566
Joined: 18 Jun 2013 23:01

Re: Linguistic pet peeves

Post by Sumelic »

It’d be really nice if there were an unambiguous technical word used in historical linguistics for the thing “cognate” doesn’t usually mean [in a historical linguistics context] but that laypeople usually think it means [due to the way the term is used in contexts like foreign language classes]. This isn’t so much a peeve, as a gap I feel in the lexicon. That is, a word that means “a word that orginates from the same morpheme/lexical item as whatever word we’re referencing/comparing it to, whether by borrowing or by inheritance from a common ancestor that the two modern languages are hypothesized to have diverged from”. It would be useful for words like “cognate” in English that are taken from Latin, and also for words in proto-languages like PIE, P-Semitic or P-Finno-Ugric that we think are related, but aren’t sure how.

So we can say “English <cat> and German <Katze> are cognate (I don’t actually know if this example is true, but it seems plausible) but English <cat> and Welsh <cath> are *glognate (come from the same root, but have been borrowed).
Or we could say, German <Katze> is cognate to the English word <cat>, but Welsh <cath> is only *glognate to it.

When I first thought about this problem, I wondered if there were biological terms that might be stolen to use for lingustics, and I actually did find some useful ones: unfortunately I haven’t seen any sign of them being adapted by the linguistics community. But if you check out the discussion here, it provides the following useful terms:

homologous - homolog - homology - “related” in the broadest possible sense: subsumes both of the relationships below.
orthologous - ortholog - orthology - inherited from a common ancestor
xenologous - xenolog - xenology - connected by borrowing

So we can say “English <cat> and German <Katze> are orthologous (I don’t actually know if this example is true, but it seems plausible) but English <cat> and Welsh <cath> are xenologous (come from the same root, but have been borrowed).
Or we could say:
The English word <cat> is homologous to the German <Katze> and Welsh <cath>.
German <Katze> is orthologous to the English word <cat>, but Welsh <cath> is only xenologous to it.

It’s best in my opinion to simply not use the word “cognate” at all in discussions with non-experts, since I’m pretty sure they are likely to have incorrect perceptions of what it means. I definitely wouldn't use “related” as a technical term for “strictly cognate”! (Some linguists actually do this!!)

The unfortunate thing though is that probably, the most confusing situation is when someone uses the word “cognate” to mean “loosely cognate” and is misunderstood to mean “strictly cognate”. In this situation, the speaker is most likely not an expert, so teaching the experts new terminology wouldn’t actually be much help in this situation. But I think the bio terms can be helpful as disambiguatory terms, and to provide a nice set of terminological distinctions.
Last edited by Sumelic on 28 Mar 2015 04:57, edited 2 times in total.
Prinsessa
runic
runic
Posts: 2647
Joined: 07 Nov 2011 14:42

Re: Linguistic pet peeves

Post by Prinsessa »

I've seen cognate used in a broader sense like that in formal papers many times.
Sumelic
greek
greek
Posts: 566
Joined: 18 Jun 2013 23:01

Re: Linguistic pet peeves

Post by Sumelic »

Hmm, that just increases my desire for a substitute for the term then; there's an even greater risk of confusion than I thought if "cognate" is regularly used in linguistics papers with the broader sense, because I've definitely read things from several historical linguists indicating that they only use it for the stricter sense.
User avatar
HinGambleGoth
sinic
sinic
Posts: 432
Joined: 01 Jul 2014 05:29
Location: gøtalandum

Re: Linguistic pet peeves

Post by HinGambleGoth »

"The swedish spelling is just as phonetic as spanish or german"

"Swedish lacks diphthongs alltogether"

"Spoken swedish is clearer and easier to pick up than danish or norweigan"

"The swedish pairs like men/män and gott/gått should be spelled the same"

"Finland swedes are bad at swedish, since they cant pronounce the sj-sound"
[:D] :se-og: :fi-al2: :swe:
[:)] :nor: :usa: :uk:
:wat: :dan: :se-sk2: :eng:
[B)] Image Image :deu:
User avatar
Xing
MVP
MVP
Posts: 4153
Joined: 22 Aug 2010 18:46

Re: Linguistic pet peeves

Post by Xing »

Sumelic wrote:Your example cuts both ways. Since [pæk] is ambiguous, rather than clearly /pæk/, this shows that aspiration is just as important as voicing for identifying voiceless aspirated stops at the start of a word.
[+1] [pæk] would probably be interpreted either /bæk/ or /pʰæk/, depending in the context and what our brains 'add' to the phonetic clues we hear. It's possibly also somewhat dialect-dependent, since many dialects of English tend de-aspirate voiceless stops.

It's like when people say that English has or hasn't length distinction in vowels - in many dialects, both length and quality help to distinguish between, say, /iː/ and /ɪ/, or /uː/ and /ʊ/, and it can depend on the listener and the context whether /i u/ are interpreted as /ɪ ʊ/ or /iː uː/.
Prinsessa
runic
runic
Posts: 2647
Joined: 07 Nov 2011 14:42

Re: Linguistic pet peeves

Post by Prinsessa »

HinGambleGoth wrote:"Swedish lacks diphthongs alltogether"
And even if that were true for the standard language that would be a very silly thing to say, as tho dialects suddenly didn't exist. :c
HinGambleGoth wrote:"Finland swedes are bad at swedish, since they cant pronounce the sj-sound"
Neither can central Swedes. c;
clawgrip
MVP
MVP
Posts: 2257
Joined: 24 Jun 2012 07:33
Location: Tokyo

Re: Linguistic pet peeves

Post by clawgrip »

I will definitely agree that aspiration is an important part of the b-p distinction. I understand why you say the /p/ from /sp/ could be /b/, but don't see any conclusive evidence though. I think some amount of native-speaker intuition is now out of line, to avoid /h~ŋ/ style phonemes.

And length I definitely agree is an important part of vowels. It's combined with stress and is kind of a complicated mess.
Prinsessa
runic
runic
Posts: 2647
Joined: 07 Nov 2011 14:42

Re: Linguistic pet peeves

Post by Prinsessa »

Now that I think of it, Sumelic, isn't the term akin (to)?
User avatar
HinGambleGoth
sinic
sinic
Posts: 432
Joined: 01 Jul 2014 05:29
Location: gøtalandum

Re: Linguistic pet peeves

Post by HinGambleGoth »

Prinsessa wrote: And even if that were true for the standard language that would be a very silly thing to say, as tho dialects suddenly didn't exist. :c
In many non-standard swedish dialects such as :se-vg: the long vowels are far purer than the central swedish schwa-centering and approximant offglide pronounciations.
Prinsessa wrote: Neither can central Swedes. c;
You are :se-vg: right? /x/ ? Older litterature described the post-alveolar sound as the standard sound, but now it is limited to northerners, finlanders and värmlanders, and Östermalm of course.
[:D] :se-og: :fi-al2: :swe:
[:)] :nor: :usa: :uk:
:wat: :dan: :se-sk2: :eng:
[B)] Image Image :deu:
Post Reply