what /p/ looks like

One of the reasons why the term “visual words” in (psycho)linguistics is so irritating is because it assumes that a written word is just the same as a spoken word except that you see it rather than hearing it.

And another thing that’s been bugging me to varying degrees recently is the assumption, all but dead in phonology but persisting even in fields which find it necessary to invoke phonology constantly, that speech consists of phonemes.

Both these points deserve (lengthy) rants in posts of their own, but mentioning them right now gives the perfect opportunity to showcase two beautiful screenshots from Praat which have been sitting around in my laptop since I created them for a presentation only to decide they weren’t needed. Behold:



(I’d show you the audio files too, except I’m too much of a cheapskate to pay £10 to WordPress for a year’s worth of the privilege of being able to upload wavs.)

What they illustrate is, 1, that the closest you get to making an auditory word “visible” is when you measure and graph its acoustic properties (and/or of course measure and graph or video its articulation) , and 2, there is very little in the way of (either acoustic or articulatory) evidence for units in the speech stream. While everyone agrees that the concept of phones, physically invariant slices of the speech stream, is naive and outdated, the view that phonemes-as-such can be identified in the speech stream-as-such still tends to linger, in spite of not only the lack of evidence in its favour but all the evidence against it. You can call incidences of approximately simultaneous bilabiality, voicelessness, and plosion “/p/” if you want, but, as can be seen from comparing the pink bits in the two waveforms and the corresponding portions of the spectrograms, this is not a conclusion which falls out of the acoustics per se.

PS – does anyone think the sections for ‘p’ are a bit long? In spite of offering the audio files I can’t actually immediately lay my hands on them to check. If it looks suspect, it will just be my segmentation at fault.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s