pronunciation of th

I cannot give you a pattern or rule for what you asked, but I can tell you myth and with both have an unvoiced th sound at the end!

That's a nice rule, not one I was familiar with.

It's not 100% accurate, though. Off the top of my head, I can think of words formed by dropping a silent "e" and adding an "ing", which retain their voiced "th"--for example, "scathing".

Intervocalic consonants tended to be voiced in some old Germanic languages (it is certainly true of Gothic), even if not written as such. This is also true of practically all modern English cases of 'th' appearing between vowels (father, wither, bother, leather). Similarly, after the letter 'r' and before a vowel, 'th' will usually be voiced. In 'north' it is unvoiced, but in 'northern' it is voiced.
That is sort of a pattern. Words of foreign origin often break the rules, such as 'mathematics' or 'atheist'. But in language, patterns are there to be broken.

'Gothic' does too, as does 'mythic, 'mythology' and the names 'Botham' and 'Gotham'.

Far more interesting is the consonontal shifting of the 'th' sounds to 'f', 'v' and 'd' sounds, most noticeable in dialects of London. De fings me brover says.

Janet makes a point: the interdental fricative in with is not necessarily voiced--it most likely depends on your dialect. But something that I wanted to add to the discussion with regards in particular to the with/myth comparison is the variable of history. Today, the words "with" and "myth" rhyme, but this was not always the case. In the English of Beowulf, "i" was pronounced /i/ (ee) and "y" was pronounced /y/ (like the u in the french word "tu"). This means that, historically speaking, the interdental fricative in "with" and "myth" in fact do not occur in the same phonetic environment. Also note petescully's fascinating observation that this consonant is voiceless when it occurs in loanwords, which the word "myth" happens to be.

English doesn't really distinguish bewtween the two sounds; not in spelling, anyway. At least around the area where I live (Plymouth), you could get away with either an unvoiced or voiced dental fricative.

One other thing about the dental fricative which is interesting (and i realise we're veering from the point a little here) is that it occurs mostly in those languages on the periphery of Europe (Greek, Spanish, English, Icelandic) but not at all in those in the centre. Wierd, huh?

And it's also found in Arabic, Persian, Hindustani (in the form of an unaspirated /t/), etc. which are in the periphery of Eurasia. Is the interdental fricative an archaic phoneme that ended up so far away from the center of PIE innovation and thus was preserved? Is this evidence for the wave model of langauge change? Or is it coincidental. If anyone has any other evidence that regards this issue, I'd be really interested to hear it.

-adam o

An interesting example, since in my dialect of English (Southern VT) "With" and "Myth" are identical when standing alone. In fact, the only time I would voice the -th in "with" is if it is followed by a voiced th- sound in the next word - "what is up wiTH THat?"

-th-

Why bother with rules if you're teaching? Sometimes they confuse the students more than they help!

Write two lists according to pronunciation (appropriate to your students' level and needs) and practise them.
You could mix them up and make a sorting game of it all.

(In certain classes (or with certain L1 students), you might want to link this to a pronunciation exercise focussing on 'voiced' and 'voiceless' consonants - e.g. dock, dog; ferry, very; wash, watch, etc..)

Enjoy your teaching!!

That's a really good idea. Rules can often mislead more than inform.
Derrick Coyne's variation is interesting, I think I have heard other forms pronounce 'with' with an unvoiced (though can't think of them now).
Does anyone know of any good books/studies/articles to read about this?

I've been noticing more and more people giving an "f" sound for "th." I've heard many educated and relatively articulate people saying booff (both), truff (truth), Ruff (Ruth), wiff (with), maff (math), etc.
Why this be?

Adam O: "And it's also found in Arabic, Persian, Hindustani (in the form of an unaspirated /t/)"

There is no dental fricative in Hindustani. There are aspirated and unaspirated dental stops, but no fricative.

Going back to the original message, has anyone else come across any rules/patterns for pronunication of "th" - I have a speech/language impaired client who is unaable to ge tthe pronunciaation right and I am looking for something by which he can guide himself. It would also help people learning English as a second language...

Some patterns have already been mentioned, but since these are 2 different phonemes, their distribution is not completely predictable.

In Old English, their distribution was predictable: voiced between vowels or before a voiced consonant, voiceless elsewhere.

Now the voiced sound is found in function words (the, that), intervocalically (mother, father), and in verbs (breathe, teethe). New words, or just non-Germanic words, usually have the voiceless sound.

either - voiced
ether - unvoiced

As for "with", I don't think it's purely a matter of dialect. I think it also depends on what comes next. I really have to think about this some more and research it as well, but I noticed that I usually make the "th" voiced if the next word starts with a voiced sound, but unvoiced if the next word starts unvoiced. For example, I would say "come wiTH me", (voiced) but would say "...with happy thoughts" (unvoiced).

There is a rule. This work sheet explains it in detail. It can be confusing at first, but eventually these "rules" will be so familiar that they are subconscious.
http://www.jamesabela.co.uk/beginner/soundpairth.pdf