Thanks, Phil, this is really super! I was hoping for a thoughtful response that would help me sort this out. Although I have a lot of experience learning and teaching languages, I’m definitely still a beginner in Mandarin, in the first few months of study.

I’ve known about the voiced/unvoiced and aspirated/unaspirated contrasts for a long time, but I just became aware of the fortis/lenis contrast in the last year. I’ll look into that; it maybe means I’ve oversimplified things in my post. But it seems to me that the following has happened.

Three sounds

1. b as in “book”, voiced, unaspirated (or lightly aspirated?)

2. p as in “pin”, unvoiced, strongly aspirated

3. p as in “spin”, unvoiced, unaspirated.

And so what has happened (exaggerated a bit to make the point), teachers/instructional materials say, “well, pinyin b should really be 3 (unvoiced, unaspirated p), but it’s too hard to get English speakers to say it that way, so let’s just tell them to do 1 (voiced, unaspirated b). It’s not quite right, but we’ll ignore the voiced/unvoiced issue in order to get it right on the aspirated/unaspirated issue.” Again, I may be oversimplifying things. And I’ll add that I’m not on some nerdy quest (although I can be nerdy :-) ) to create the perfect chart, or prove that I’m right; I just want to get the best Mandarin pronunciation I can. And to that end, it seems unhelpful to not get it right with voiced/unvoiced. Of course, as you point out, maybe that’s not where I should invest the most effort.

Please explain “layering prosodic intonation on top of the lexical tones”.

And yes, please point me in the direction of that italki teacher. BTW, I love italki. I had a fantastic experience with a Romanian teacher on italki. For Mandarin, I’ve decided to take a local course in Portland, Oregon.

Part 1

Yes, Mandarin (and Cantonese) consonants other than l, m, n, and r are all phonemically unvoiced. Traditionally, this was considered a big deal, since the Europeans who dealt the most with China were Slavic and Romance language speakers. That’s why English translates 道教 as Taoism (with a T) to this day. And the P in Taipei is the same sound as the B in Beijing. Older romanization schemes wrote stuff like P and P’, but people never bothered to write the apostrophe (actually, there’s another word for it but I forget), so it was hard to read. The modern systems use P and B, which more accurately reflects the aspirated and unaspirated sounds’ status as separate phonemes.

Native Chinese and English speaker here. I'm not very versed in linguistic terms. Does voiced refer to the vibration that starts before speaking the word, kind of like "mm-book" or "nn-duck" or "nn-get"? Because that seems to sometimes not occur, especially in faster speech, in which case the consonant sounds prettyt much identical to its pinyin equivalent in Chinese to me.

Part 2 of 2

However, the reality is that English “voiced” consonants are not truly fully voiced. In reality, English has a fortis / lenis distinction that combines unvoiced / voiced with aspirated unaspirated, depending on the syllable structure and stress. (The same is true with standard German, but not with Romance or Slavic languages.) It should also be noted that in standard Mandarin accent, unaspirated consonants will indeed be fully voiced in neutral-tone syllables — it’s just a lot easier. By the way, I work with a lot of Chinese students who are learning English (some actually have excellent pronunciation), and no matter how bad their English pronunciation may be, it is never ever the voiced / unvoiced, aspirated / unaspirated distinctions that give them problems with the pronunciation of English consonants. 

If you’re really interested, I can point you in the direction of an italki teacher who is really a stickler for using unvoiced consonants in Chinese — just let me know. But it’s really not a big deal compared to such points as getting your tones right and layering prosodic intonation on top of the lexical tones. One should also pay attention to the different sibilant series (by place of articulation), but with over a billion speakers, there are actually hundreds of millions who never distinguish between zh ch sh and z c s. And in southern China, most speakers merge initial L and N. Really, the best thing (for Westerners in particular) is to prioritize tones.