47

Having practiced armchair linguistics for some years I should be able to sum up the difference off the top of my head, yet often I don't know which term to use.

And looking them up on Wikipedia doesn't help a lot...

Wikipedia on phonology:

Phonology is, broadly speaking, the subdiscipline of linguistics concerned with "the sounds of language".

Wikipedia on phonetics:

Phonetics is a branch of linguistics that comprises the study of the sounds of human speech.

Can it be that the difference is that phonology deals with language sounds and phonetics deals with human speech sounds? And if so, well what does that mean?

3
  • 1
    You can consider this a test case of a naive questioner asking expert answerers, but I really do get them confused. Sep 15, 2011 at 22:47
  • 15
    Here's a humorous illustration I sometimes use in introductory courses: specgram.com/CLIII.1/09.parenchyma.cartoon.e.html
    – user652
    Jan 6, 2012 at 19:26
  • 2
    The two have different histories of study, but in my understanding, there is no theoretical distinction between phonology and phonetics. Look though you will, I don't think you'll ever find a satisfying answer to your question.
    – Greg Lee
    Apr 29, 2020 at 13:31

15 Answers 15

40

Phonetics is about the physical aspect of sounds.

In phonetics, sounds are called phones.

Phonetics has subcategories where it studies different kinds of sounds.

But in general, we usually mean articulatory phonetics: the study of the production of speech sounds, by the articulatory and vocal tract of a speaker, and also their perception.

Phonetic transcriptions are done using square brackets like these: [ ].


Phonology studies the abstract aspect of sounds.

In phonology, sounds are called phonemes.

Phonology is about establishing what are the phonemes in a given language, where a “phoneme” is defined as a sound that brings a difference in the meaning of a word.

Consider the following minimal pairs, in which a change in sound causes a change in word, and meaning:

  1. bat vs. pat
  2. had vs. hat

This example is in Italian:

  1. pèsca (-> /ɛ/) means “peach”
  2. pésca -> /e/ means “fishing”

Phonemic transcriptions adopt the slash, like this: / /.


A phoneme is a “phonic segment” - a unit from phonetics - plus a linguistic “meaning value”.

4
  • 4
    Ah I think one thing that keeps me confusing them is terms like "phonetic transcription" which often deal in phonemes but phonemes belong to phonology and not phonetics \-: Sep 15, 2011 at 23:30
  • 2
    I added that part to the answer, it looked better there. :) By the way, I think this is one of the things that every student studying Linguistics have wondered about.
    – Alenanno
    Sep 15, 2011 at 23:48
  • 2
    Actually, a phonetic transcription should just deal with "phones" not "phonemes" - which belong to phonemic transcription. But that probably doesn't make it too much clearer!
    – LaurenG
    Sep 16, 2011 at 1:01
  • 3
    Yes when people want to contrast phonetic and phonemic transcriptions they use the right term. But otherwise phonetic transcription is naively used to cover both so it's more vague or ambiguous. Think phrasebooks, non-linguists talking about languages in online forums etc. Sep 18, 2011 at 8:31
26

My advisor, Dennis Preston, used to tell students that the ear hears phonetics, but the brain hears phonology. That is, your ear is capable of processing whatever linguistic sounds are given to it (assuming someone with normal hearing), but your language experience causes your brain to filter out only those sound patterns that are important to your language(s).

Of course, this summary simplifies things considerably. Phonologists are often as interested in patterns related to the manner of articulation as they are the patterns of the speech waves. Phoneticians, meanwhile, would have no way to analyze their data sets if they didn't have phonological categories to help organize them.

Generally, phonetics is the study of fine grained details of those sounds, while phonology has traditionally dealt with analysis of greater abstractions. For understandable reasons, the line between the two discipliens is blurring, particularly as our modeling capabilities become more sophisticated. Still, the distinction is useful.

16

I think the big difficulty with the phonetics-phonology divide is not only that linguists don't even really agree on the difference but also that there doesn't exist a good analogy with any other pair of subfields.

This is the way I've seen it (cards on the table, although there are more extreme folks, I'm fairly far on the "phonology doesn't exist" camp, and that is probably influencing my answers)...

Phonology is the study of the cognitive processes that turn words into instructions to hand down to the physical body parts that produce the sounds. These instructions, personified into human commands, might sound like, "close your lips, now move your tongue to touch your alveolar ridge; begin lowering the diaphragm at a normal rate and constrict the vocal chords to this degree". On the acoustic side, phonology's role is much harder to specify (at least to me), but I would say that the "phonology" center takes in sequences/matricies of interpreted linguistic features, for example "between 442-488ms, palatalization level 2". Phonology would then turn that into the abstract "underlying" representations that can be mapped to morphological parsers and the lexicon.

Phonetics is the study of how the "commands" end up translating into specific articulator and vocal tract movements. For instance, how the command to retract the tongue at some particular time "really" maps to minute physical details like exactly when tongue section X touches mouth section Y and then in turn how that affects parts of the resultant acoustic signal. Phonetics also makes observations of how certain groups of instructions can cause very specific consequences. On the acoustic side, phonetics turns the mental spectrogram we receive from the nerve endings in our cochleas into feature sets and timings of the sort that it received from the phonological center during articulation.

Articulatory phonology is an attempt to consolidate the two, that, as far as I can tell, is basically phonetics taken one level deeper to receive underlying segments as inputs. And articulatory phonology moves a lot of what was in phonology proper as cognitive processes into physically motivated processes during articulation.

In short, nobody really knows the difference, but the broad agreement is that phonetics is lower-level and more articulator-centric and phonology is higher-level and more cognition-centric.

12

Firstly it should be pointed out that there is some overlap between these two subdisciplines of linguistics, just as there is overlap between, say, syntax and morphology. But you're not far from the right track when you say:

"the difference is that phonology deals with language sounds and phonetics deals with human speech sounds..."

This is close, but it doesn't encapsulate the distinction memorably. I suggest the following approximation:

  • Phonology: how sounds pattern within a given language (stated in terms of "phonemes")
  • Phonetics: the characteristics of speech sounds (stated with descriptions of speech those sounds, sometimes referred to as "phones") themselves

Crucially, a phoneme consists of a set of phones, plus a set of rules describing how those phones are distributed within a particular language.

So, if we refer to a "voiced palatal affricate" [dʒ] without context, we are describing a "phone" -- a speech sound which is produced through a particular combination of articulations.

However, if we make a statement like "In Japanese, the phoneme /d/ has the allophone [dʒ] before the vowel /i/," then we are describing the patterning of phones in Japanese. Thus, this is a phonological description.

Note that how a particular phone patterns may be different in another language.

In English, for instance, "deep" [diːp] and "Jeep" [dʒiːp] are distinct terms, so while the phones in question here are comparable (roughly) to those in the Japanese case, the patterning is different.

However, and this is where it gets a bit messy, it's usually the case that there are purely phonetic (articulatory or auditory) influences that "motivate" particular phonological distributions. In fact the example from Japanese above is of a sort that's quite common cross-linguistically, so much so that it's been given its own name, "palatalization." So, phonology can often be "explained" in terms of phonetics.

Both of these subdisciplines have huge literatures and of course I'm grossly oversimplifying, but hopefully that is a useful start.

2
  • Is ぢ still rendered /di/? (I know it is in Nihon-shiki romanization). Then how about ディ, which is pronounced differently? There has been a phonemic split.
    – dainichi
    Jan 29, 2012 at 9:17
  • 1
    This is the first correct answer I've read yet—thanks! Feb 4, 2014 at 0:18
10

Phonetics deals with sounds. Phonology deals with phonemes. What is a phoneme? Ah, that's one of the things phonology deals with. Nikolai Trubetskoy defined it as “the smallest distinctive unit within the structure of a given language”. Now, what does that mean?

For example in English if someone says [riŋ], you will understand ring, even if you pronounce it [ɹiŋ]. So we can consider that the differences between [r] and [ɹ] are not distinctive. Thus, we call [r] and [ɹ] unbound variants of a phoneme /r/.
Now if someone says [piŋ] it has a different meaning : it's ping. So [p] is not a realisation of /r/, but rather of another phoneme /p/.

This process is the most obvious way of finding the phonemes of an arbitrary language, and [riŋ] and [piŋ] what we call a minimal pair. A pair of words different only for one sound, but with distinct meanings.

As you have probably guessed it's a bit more complicated, but that's the bulk of it.

Now why bother with phonemes? Why aren't sounds relevant enough?

It is a matter of abstraction, convenience. Though interesting, phonetic systems of languages often lack obvious patterns. Studying them is tedious, and subject to bias. And there are many topics in linguistics that don't demand such precision.

Phonological systems, though present many regularities, are far easier to formalise, and in fields such as morphology, semantics or pragmatics, they are just precise enough. There is no need to know if one says ring [riŋ] or [ɹiŋ] if they bear the same meaning.

So there are differences, these two fields don't address the same issues, though they are closely related. And both are useful.

3
  • Most of the literature purporting to be in phonology deals with subphonemic phonetic details. So I think your answer is wrong.
    – Greg Lee
    Apr 29, 2020 at 13:47
  • @GregLee Yes, the question of what are the phonemes in a language, what are the concrete realisations of a phoneme is central in phonology. My point is not that phonology doesn't deal with sound, but rather that phonetics doesn't deal with phonemes, maybe I should make it clear but this was 9 years ago…
    – Evpok
    Apr 29, 2020 at 15:13
  • But if you just sample the phonetics literature, you'll see that phoneticians talk about about phonemes all the time.
    – Greg Lee
    Apr 29, 2020 at 21:43
6

A simple way to show the difference is that

  • phonetics is the study of possible sounds that a human mouth can make and human ear can hear over all languages
  • phonology is the study of those same sounds within individual languages (which in each language is a much smaller set that than the total possible).

When studying a particular language, phonology is mostly all that matters. Comparison between languages brings out the phonetics. Catalan has one nasal phoneme (doesn't distinguish m,n, and ng, even though they are pronounced as such in different contexts). And English doesn't distinguish some things that make semantic differences in other languages.

6

An additional answer would be that phonetics is concerned with the physical processes of speech (production, processing) while phonology is concerned with abstractly modeling speech processes. There is a lot of overlap between the two areas and it is difficult at times to figure out if a process should be considered phonetic (a physical result) or phonological (basically a rule that is not physically necessary).

An example of this is something like voicing assimilation. Say you have two words which in isolation sound like:

[ap] [ga]

And you put them together:

/ap+ga/

And the resulting sound is:

[abga]

The [p] became [b]. Why? A phonological answer could be that /g/ has some feature controlling its voicing [+voice] and /p/ has a feature [-voice] making it voiceless. Then putting them adjacent, the /p/ latches on to /g/'s [+voice] feature and let's go of it's [-voice] feature, making it a voiced /p/ ([b]).

A phonetic answer could be something like, in anticipation of the voicing for producing [g] you also voice somewhat during a production of [p], making it more [b]-like.

2
  • But what do you say about [abga] if you also find a pronunciation with simultaneous labial and velar closure, instead of first labial closure only followed by velar closure only. Is this a phonological fact, or a phonetic fact?
    – Greg Lee
    Apr 29, 2020 at 14:02
  • @GregLee -- good question. If there's a simultaneous closure, then there's a simultaneous closure, it's a phonetic "fact". That said, it's just one data point. It's like how in phonology we say a phoneme is voiced or unvoiced, but looking at spectrograms you often see incomplete voicing across the duration of an articulatory gesture. Apr 30, 2020 at 17:28
5

This is an example I always find is helpful for first year. Ignore the vowels just for now, and just focus on the 'p' sounds. If I were doing a phonetic transcription of English notice how all the 'p' sounds are different and they've got slightly different diacritics to show this:

[pʰin] 'pin' - notice it's aspirated! [spin] 'spin' - notice it's not aspriated! [stop̚] 'stop' - notice that you didn't actually open your mouth at the end.

I use those square brackets to show it's phonetic - it's the sounds as they really and objectively are. But that's not how English speakers really think of them - they don't think of these sounds being different, they all just think of them as 'p.' to show this we just write them all as /p/ and use angle brackets to show that people think of them all as the one sound - /pin/, /spin/ and /stop/.

1
  • Yes I've always been clear on phonetic vs phonemic transcriptions, just not on phonetics vs phonology. Also, many places don't seem to use /this/ and [this] consistently which doesn't help. Sep 16, 2011 at 1:12
5

I'm surprised that nobody has mentioned the form/function distinction.

Phonetics studies the nature (acoustic and articulatory) of sounds that humans produce while speaking. Phonology studies their function in differentiating meaning in various contexts.

Thus (very crudely), phonetics will notice the difference between the aspirated and non-aspirated /p/ in pin vs. spin. But phonology will be mostly interested in the fact that the meaning of both words will change when you replace /p/ with /k/, in exactly the same way as in lip vs. lick where aspiration doesn't enter the picture.

But of course, phonology can only be successful when it has good phonetics to rely on.

4

This Reddit comment chain helped me:


kjoonlee 5 years ago [as of 2018 Mar. 1]

Short simple answer:

Phonetics deals with sounds as they are made in your throat/mouth/nose/etc. and sounds as they are received by your ears. (Articulatory & acoustic phonetics vs. Auditory phonetics) This also means phoneticists/phoneticians can learn to become speech therapists and dialect coaches.

Phonology deals with sounds as your brain perceives them. And there is a difference between the ideas of sounds (phonemes) and the actual acoustic sounds (phones/allophones) themselves.

For example, you might think you say the same "t" sounds in "tar" and "star" but they're actually different acoustically.


lalalalalalala71 5 years ago

This post is the best answer.

For example, you might think you say the same "t"sounds in "tar" and "star" but they're actually different acoustically.

So, OP: phonetics is the discipline which tells you the /t/ sounds in tar and star are different (in tar, it is "aspirated"; that is, it is pronounced along with a puff of air, like a very soft version of the sound of the letter H. This aspiration is absent in star). Phonology, on the other hand, is the discipline which tells you these sounds are the same, as they're just variations of the sound /t/.

3

It's useful to think of phonology as more than just the individual sounds that are distinctive in a given language, and instead as the 'grammar' of those sounds. A phonological description should not just be 'here are the vowels and consonants I found forming contrasts in language X' - it should also tell you about the rules that govern these sounds, such as what sorts of combinations of sounds are possible (phonotactics), what the stress patterns are and/or what the patterns of tonal marking are, relevant prosodic characteristics (e.g. in many languages a question is marked by higher pitch utterance-finally), and of course, the allophonic variation that is permitted for each phoneme, and in what sorts of environments certain allophones occur. Phonetics, on the other hand, studies the acoustic and articulatory substance of speech data, and the techniques used by phoneticians offer a great deal of detail about the sounds of a language. This helps phonologists to back up their claims about different types of sounds (or maybe demonstrates that someone's impressionistic observations are not supported by the actual phonetic data, as is sometimes the case). Phonetic analysis also helps to explain why certain phonological patterns occur, because it is possible to analyse how and to what extent different sounds affect each other, and these phonetic effects are important factors in allophonic variation and language change. The two are not mutually exclusive and there are many linguists who work in both phonetics and phonology, because these two sub-fields are, of course, closely related, but also dependent on each other.

3

Phonetics is what comes out of your mouth and goes into your ears. Phonology is how those sounds are arranged in your head to go to your mouth, and how the sounds from your ears are deconstructed to be processed by your brain.

2
  • This seems like a great summing up to keep the distinction clear, but it implies both fields deal only with production and neither with perception. May 20, 2013 at 2:10
  • 2
    Slightly less succinct, but it covers both now! May 20, 2013 at 5:35
1

phonetics is the production and perception of speech sounds in any language and it deals with "phone". Phonology on the other hand is the interpretation of speech sounds in a particular language and it deals with phoneme: the smallest unit of sound.

2
  • It's a good point that you include perception in phonetics along with production I think. But with the concept of the phoneme being largely discredited wouldn't this mean that phonology has also been largely discredited, which I don't think is true. May 20, 2013 at 2:12
  • @hippietrail I think there's a big difference between disputed and discredited. Jun 17, 2016 at 10:21
1

There's no difference in substance, but phonologists love to make theories and phoneticians hate it.

2
  • 3
    Phoneticians love theories too. Just different types of theories. Jun 17, 2016 at 10:27
  • 4
    @Araucaria, Please name some theories loved by phoneticians. I can only think of Fant's acoustic theory of speech production and Delattre's articulatory theory of speech perception. Theories make predictions.
    – Greg Lee
    Jun 17, 2016 at 18:06
0

Phonology comprises language-specific rules about which sounds a language has, how they can be combined, and how they change in various phonetic contexts in a particular language.

Phonetics is the physical description of and explanation for human speech sounds no matter what language happens to use them. There are two kinds of phonetics: articulatory, which describes sound in terms of where and how they are made, and acoustic, which describes sounds in terms of their acoustic properties, as revealed on sound spectrographs.

2

Not the answer you're looking for? Browse other questions tagged or ask your own question.