Jj is for Jottings 60.  Partial Devoicing of Consonants.

As I said in the previous article “most voiced consonant sounds on the ends of words are actually devoiced , which means they end up sort of halfway between being voiced and unvoiced.”  It would really be more accurate to say that they are partially devoiced.  In order to really get a handle on this you will actually have to say words and listen to how they sound and feel them, so make sure you are on your own or in a place where people won’t think you’ve gone bonkers.

Say these two sentences, and compare how you say the /g/ on the end of “dog”:

I saw a dog and a cat.

I saw a dog.

You’ll find that you say the /g/ fully and completely in the first sentence because the /g/ is followed by a vowel.  It is as if the /g/ is not on the end but in the middle (which it actually is, in a way).  In the second sentence the /g/ is truly on the end and you will find that you have not given full voice to the /g/ – you have your articulators in /g/ position (back of tongue up against soft palate), and you bring your vocal cords together for voicing, but not much more.  But it clearly isn’t a /k/ sound: it’s somewhere between /k/ and /g/.  This is also true when the final voiced consonant is followed by a voiceless consonant eg. “dog toy”.

You can play around with this for any voiced consonant and you will see that the same thing applies.  For a list of voiced consonants see here.

When a child is sounding out a word, there sometimes need to be small adjustments made between sounding out each sound and the final product.  Continuing with the “dog” example, when saying each sound the child will naturally – and quite correctly – say a fully voiced /g/ on the end, but when blending it into a natural-sounding word, less voice is needed for the /g/ because it is at the end.