I am from the Czech Republic, and one thing that has always bothered me is that a lot of English (and other) loanwords were written in Czech with "dž" in place of the English J, e.g. "jeans" -> "džíny". I always thought that the "j" and "dž" sounded completely different but I assumed that the Czech alphabet simply doesn't have the appropriate consonants to transcribe it properly. But later, to my surprise, I found that 1. this "dž" combination is actually very common in many other languages and that 2. even the IPA for the English "j" sound is /dʒ/ (which is equivalent to dž).

However, recently I realized that there is a different pair of consonants that actually sounds identical to the English "J" sound, and even if I'm maybe just unable to hear the difference, I am 100% sure that it is at least much closer to it than "DŽ". As you already know from the title, that pair is "ĎŽ", or /ɟʒ/ in IPA.

So why exactly is "dž" or /dʒ/ used for this sound rather than "ďž" or /ɟʒ/, which is much closer? Is there an actual reason for it or is it just an oversight due to the western linguists' unfamiliarity with the standalone "Ď" /ɟ/ sound?

  • 4
    The English post-alveolar affricate definitely does not begin with a palatal plosive.
    – Graham H.
    Commented Dec 8, 2023 at 14:18
  • I just felt the need to point out that jeans in Russian is джинсы; and д corresponds exactly to Czech d, and ж corresponds exactly to Czech ž. What's more, Lithuanian, Yiddish, Kazakh etc. etc. all do something very similar to transcribe this sound. Commented Dec 8, 2023 at 14:35
  • @GrahamH. It sure doesn't begin with a D either.
    – Detheroc
    Commented Dec 8, 2023 at 16:32
  • 2
    @Detheroc [d] can refer to multiple things depending on how narrow the transcription is. I'm not sure, but many languages (including some people's Czech), have dentialveolar, or even purely dental d, whereas this is quite unusual in English, which tends to be true alveolar d. So what may be transcribed [d] in English need not be identical to the sound transcribed [d] in Czech
    – Tristan
    Commented Dec 8, 2023 at 16:38
  • 1
    This is basically the same issue as the pronunciation of "new". Czech speakers might swear up and down that it begins with a "ň" / ɲ, and English speakers might swear up and down that it begins with an ordinary n.
    – Pilcrow
    Commented Dec 8, 2023 at 23:11

3 Answers 3


A sound in a language can "actually" be two things. One is, what it actually is phonologically. The other is, what it is phonetically, in this case, exactly now is it articulated? Phonologically in English, it is a single segment. In Czech, it could be a cluster, or it could be a postalveolar affricate (single segment), I don't know of any compelling arguments one way or the other. In the APA that affricate is transcribed as [ǰ], and in the IPA it is (usually) transcribed as [dʒ]. The APA transcription recognizes the unitary nature of the affricate, but the IPA does not recognize the concept "affricate", so you are forced to use a two-symbol sequence. The letter <d> in IPA is ambiguous, standing for any of dental, alveolar and postalveolar. If it is necessary to mark a "t" as dental (as is the case in some Dravidian and Australian languages because there is a dental / alveolar contrast) you can us a diacritic as in [t̪]. There is no symbol or diacritic for a postalveolar stop. The writing <dʒ> does not say that the closure portion is different from the release portion, and in fact there is no unambiguous way to say – using symbol choice – that the two phases of the consonant are at the same place of articulation. If you want to indicate "a unit", you can write <d͡ʒ>, which can be necessary in some languages (when clusters and affricates contrast).

When you transcribe words of a language in IPA, you might only be transcribing phonemically, in which case you only use the minimal symbols for phonemes of the language. In Polish, you have to transcribe trzy "3" differently from czy, where the former has a cluster and the latter has a single consonant affricate, which is where the tie diacritic is pressed into service. As far as I know, Czech does not face that problem. Alternative transcriptions like [ďž] simply are not IPA, so a better question would be, if you're not using IPA, why don't you follow the APA standard and write [ǰ]. The IPA-consistent transcription [ɟʒ] is rejected for two reasons. First, there is no phoneme [ɟ] in Czech. Second, [dʒ] is auditorily closer to the phonetic output than [ɟʒ] is, and the general rule is to pick the closest symbol, consistent with the extant phoneme symbols. The alternative [ɟ] is rejected for "dʒ" because that stands for a completely separate phoneme (the one spelled <ď>).

Sticking only to symbols with phonemic status in Czech, the cluster transcriptions [dɟ, dʒ] as phonetic descriptions both suffer from the same problem of not explicitly saying that the closure and release are at the same place of articulation, you have to add a footnote saying that. [dʒ] is better than [dɟ] given the standard of auditory closeness.

  • 1. My question is not about Czech, it is about how the sound, which is written as "J" in English, is represented in IPA and in various languages internationally. 2. My central point was that [ɟʒ] actually is phonetically closer than [dʒ] and I can very clearly hear that. You just say that it isn't without any supporting evidence.
    – Detheroc
    Commented Dec 8, 2023 at 16:30
  • 3
    @Detheroc how you hear it is strongly influenced by your native language. Your perception is not good evidence that English j is actually closer to [ɟʒ] than it is to [dʒ]
    – Tristan
    Commented Dec 8, 2023 at 16:40
  • @Tristan I do realize that and I thought this might be the case initially but no matter how long I tried, I wasn't able to pronounce d and ʒ in a way that would actually resemble the English J, but if I do the same with ɟ (ď) and ʒ it immediately sounds similar. Additionaly, I've heard J pronounced like Ď in some American accents which can't be an accident, right?
    – Detheroc
    Commented Dec 8, 2023 at 16:46
  • Polish also has that contrast between voiced affricate and plosive+fricative cluster, e.g. dżem ("jam (food)") and drzem (imperative of drzemać "to doze").
    – Arfrever
    Commented Dec 9, 2023 at 3:50

I think your question can be rephrased as "why is the ʤ sound in English typically transliterated into Czech as dž and not ďž?" Since I don't know Czech, I can't comment on which option is "closer" to the English sound. But some possible reasons might be:

  1. Sound change in Czech. Maybe when the earliest transliterations were happening, the pronunciation of Czech affricates was different than it was today.
  2. Orthographic founder effect. Maybe an influential scholar decided that dž was better, and everyone else just copied them without really thinking hard about it.
  3. Native orthographic neighborhoods. Maybe in written Czech it is really common to see ďž sequences and really rare to see dž sequences, so the transliterators chose to use the rare sequence to make the word look more foreign. Or alternatively, they might have chosen the less rare sequence to make the word look more natural.

For what it's worth, phonetically it is very common for the closed and open parts of an affricate sound to be homorganic (having the same place of constriction) when they're pronounced, regardless of what symbol(s) are chosen to represent the affricate. The affricate sound of "j" in English is homorganic in almost all cases I've encountered. If you want to represent that in IPA, you could use d̠ʒ (d with "combining minus below" diacritic, meaning the closure is "retracted d" AKA "postalveolar d") --- strictly speaking this should also get a tiebar like this: d̠͡ʒ. Alternatively you can use the precomposed digraph ʤ. To me those all convey the idea that the affricate is postalveolar.

It's probably safe to assume for any language that coronal affricates are homorganic unless there is strong evidence / an explicit statement to the contrary --- here strong evidence means articulatory or acoustic analysis by a linguist, not the transcriptional conventions of a particular language's orthography.

  • Maybe I phrased my question poorly since everyone thinks I'm asking about why it's written that way specifically in Czech. That is not my point at all, I am wondering why this dž / dʒ / дж sequence is ubiquitous in many languages and especially the IPA. Regarding your third point, dž is only present in loanwords and ďž doesn't appear in Czech at all - it is entirely "my invention".
    – Detheroc
    Commented Dec 9, 2023 at 21:53
  • So you're asking "why do so many transcription systems (including IPA) use a d glyph (or its language-specific analogue like д) to represent the closure portion of the postalveolar affricate"? If so please edit the question accordingly, or perhaps start a new question.
    – drammock
    Commented Dec 11, 2023 at 16:08
  • It's not about the "glyph" they use, it's about whether or not the "closure portion of the postalveolar affricate" actually is a /d/ or a /ɟ/.
    – Detheroc
    Commented Dec 11, 2023 at 22:58
  • as I said in my answer, it's typically postalveolar. I.e., d-with-retraction-diacritic. This is different from both [d] and [ɟ]
    – drammock
    Commented Dec 12, 2023 at 15:20

To add on to what user6726 said, I'll give a summarized, concise answer.

You're half right! English "j" is not [dʒ]. But, that's because it's actually [d͡ʒ].

You're probably trying to pronounce [d] and then [ʒ] when you're pronouncing [dʒ]. That's indeed how [dʒ] would be pronounced, but this is not the same as [d͡ʒ]. [d͡ʒ] represents a single sound where the position of the tongue is ia hybrid of those for [d] and [ʒ].

  • Is it then possible that this hybrid position actually results in something that is closer to the position of /ɟ/ rather than /d/?
    – Detheroc
    Commented Dec 9, 2023 at 21:47
  • 1
    @Detheroc I'm afraid not. When pronouncing [d͡ʒ], the (moving) position of the front of the tongue is the same as that of [d], and the position of the back of the tongue is the same as that of [ʒ]. [d] and [ɟ] have very different positions. That said, [ɟʒ] does sound a lot more like [d͡ʒ] than [dʒ], so I can see where the confusion comes from.
    – user43244
    Commented Dec 9, 2023 at 23:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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