Distinguishing the "th" sounds in a spelling refor

Guest   Fri May 18, 2007 7:29 pm GMT
"In Fanetiks, you have "th" for the voiced sound in "that", and "tth" for the voiceless sound in "think". I think "dh" for the voiced sound in "that" and "th" for the voiceless sound in "think" would be better, because "tth" for the sound in "think" is extremely cumbersome and lengthy and causes a whole bunch of words originally starting with "th" to be lengthened. "tthing", "TTherzdae" etc. would likely be resisted due to their length. Using "dh" for "that" and "th" for "think" doesn't give the length problem."

Response: "DH is the kind of silliness that discredits spelling reform. There is not a SINGLE ordinary person who would for so much as ONE SECOND think that DH is a reasonable replacement for TH. It looks like Brooklynese. Cheers."

How do you think a spelling reform should deal with the two "th" sounds assuming one happens? I've seen different suggestions.
Travis   Fri May 18, 2007 7:36 pm GMT
I myself would just go for "th" for /T/ and "dh" for /D/. As for "tth", I have to say that just looks horribly unwieldy and ugly, at least in my own opinion.

Anyways, if one is to significantly reforming English orthography one should probably distinguish these two, so simply leaving them both as "th" is not really the best thing to do. The matter is that even though the two are generally predictable in word-initial positions in practice due to varying by word class (grammar words generally having /D/ and other words generally having /T/, due to grammar words having been more susceptible to sandhi voicing than other words), the two are not predictable in medial positions at all (even though there seems to be a tendency for older words in English to have /D/ intervocalically and /T/ in other medial positions), and there are minimal pairs for /T/ and /D/ word-finally (generally noun-verb pairs) and some anomalous cases of word-final /D/ such as "with" in many dialects (even though the vast majority of words that are not verbs have /T/ rather than /D/ word-finally).
Adam   Fri May 18, 2007 8:26 pm GMT
Old English had two separate letters for the two TH sounds.

One of them was , (eth). This represented the "th" sound in words such as "them".

The other was , (thorn or orn) representhing the "th" sound as in the word "think."

Once upon a time, English - like other Germanic languages - wasn't written in the Roman alphabet as it is today (the Romance languages have always used the Roman alphabet).

Old English was originally written in futhorc, an alphabet whose characters are straight lines (often a combination of several straight lines) drawn at various angles. This alphabet is more commonly known as the runic alphabet. The name futhorc is an acronym formed from the first six letters of the 29- to 33-character Anglo-Saxon runic alphabet: feoh, ur, thorn, os, rad, and cen.

When the Roman alphabet was adopted by the English language, there were a few problems and a few wrinkles that took a while to iron out.

For instance, there was no letter in the Roman alphabet for the sounds now spelled with "th". Thus, thorn and eth continued in use until about 1500, by which time they were fully replaced by th. This was probably because letters combination "th" was easier to print using the new-fangled printing pressed than eth and thorn were. (The consonant combination sh had replaced sc for the /sh/ sound a bit earlier, and the history of ch is unclear).

Eth had come to look like a y, which is why the sometimes appears as ye, especially in archaic phrases or titles such as ye olde shoppe. For this reason, ye should be pronounced the same way we pronounce the.

Another problem was that there were no separate letters for the consonant j and the vowel i or for the consonant w and the vowel u. The differentiation came much later.

People who agree with a spelling reform reform for the two "th" sounds come advocate bringing back eth and thorn.

After all, Icelandic still has them.
Travis   Fri May 18, 2007 9:02 pm GMT
One note, though, is that Old English did not actually have a phonemic distinction between [T] and [D], the two being allophones, with the latter occuring for /T/ between vowels and or voiced consonants. One must remember that the use of the thorn and the eth in written Old English actually was phonetic rather than phonemic. However, it did have the two phonemes /T/ and /T:/, with /T:/ notably not undergoing voicing to *[D:].
furrykef   Fri May 18, 2007 9:32 pm GMT
Honestly, I think the lack of distinction between voiced and unvoiced "th" is one of the least significant deficiencies of our spelling system.
Travis   Fri May 18, 2007 9:51 pm GMT
It is more a matter of completeness - if one is to reform much of the rest of English orthography, why not reform this as well at the same time?

It iz moor e matter ov kumplytnis - if wun iz tu ryfoarm mutj ov dhe rest ov ingglisjoarthograffi, hwai not ryfoarm dhis az wel at dhe seem taim? (Sorri foar dhe lak ov daiekritiks hyr.)
Guest   Sat May 19, 2007 2:51 pm GMT
English looks silly like that.
It looks Germanic... :X
(i know it is)
Travis   Sat May 19, 2007 5:08 pm GMT
>>It looks Germanic... :X<<

That was sort of intentional, I have to say.

At the same time, it is not really a clone of any orthographic style of any existing Germanic language, having been influenced by Dutch orthography but at the same time having been influenced in some ways by Old English and Middle English writing, Standard German orthography, North Germanic orthographies, and even some things such as West Frisian orthography, writing in Alemannic dialects, Afrikaans orthography, and even Old Frisian writing. At the same time, it has a number of elements which are more ad hoc which were added just to fit the needs of English (such as distinguishing /s/ and /z/ wordfinally with "s" and "z").
Lazar   Sat May 19, 2007 7:05 pm GMT
<<DH is the kind of silliness that discredits spelling reform. There is not a SINGLE ordinary person who would for so much as ONE SECOND think that DH is a reasonable replacement for TH. It looks like Brooklynese. Cheers.>>

Just another bombastic and patronizing response from that idiot Schoonmaker.
Liz   Sat May 19, 2007 7:23 pm GMT
<<Just another bombastic and patronizing response from that idiot Schoonmaker.>>

Ever got the feeling that Guest is the same bloke as the troll who started that stupid "My pronunciatiation system is the best one" thread?
Liz   Sat May 19, 2007 7:26 pm GMT
Sorry...no. It wasn't Guest who wrote this nonsense. He's just asking questions.

I retract what I said.
Jim   Sun May 20, 2007 4:19 pm GMT
It all depends on yor flaver ov spelling reform. If we'r going for minimal reform tu nun at all, then we can get by with <th> for both fonemes. If, on the other hand, we'r being much mor thurra (thoro), then completeness wood suggest that we du make this distinxion.

We then cum tu the alfabet question. Du we stik tu the twenty-six reguler letters? Du we ad new letters (or ad old wuns bak)? Du we use diacritical marks? Du we scrap the old alfabet in faver ov a new wun?

If we'r extending the alfabet, eth and thorn wood be good choices but if we'r sticking tu the twenty-six, I'd prefer <th> vs. <dh>. <Tth> is terrible. Yep, long, cumbersum and ugly and, unlike with <th> vs. <dh>, there's no clu as tu which ov <th> and <tth> is ment tu be the voiced wun.

Just tu clarify: thorn is a rune but eth is a modified <d>, also it was thorn which came tu look like <y> giving rise tu the likes ov "Ye Olde Shoppe".
Lazar   Sun May 20, 2007 4:49 pm GMT
I think that if we have a spelling reform, it should simply ignore the /T/ - /D/ distinction, and use "th" in all cases. This distinction isn't shown in writing today, its distribution is largely predictable, and it carries basically no functional load.
Josh Lalonde   Mon May 21, 2007 12:36 am GMT
It depends of course on what the goal of the reform is. A phonemic system is impossible cross-dialectically, but a "most/oldest distinctions" system like Travis's is possible (though it reduces the benefits of the reform for those of us with a lot of mergers). Wells's proposal that we discussed a few weeks ago uses a "minimum distinction" system that I think goes too far the other way; in stereotypical Canadian fashion, I prefer something in the middle. The /T/ vs /D/ distinction seems pretty safe to drop to me. The only minimal pair I can think of is 'ether' 'either' and that only works for those who have the FLEECE vowel in the latter. Wells also suggests merging GOOSE-FOOT-STRUT in spelling. I have to admit that this is odd for me, though the functional load for the FOOT set is pretty low. 'Put' vs 'putt' is the only FOOT-STRUT minimal pair I can think of, and there are few for FOOT-GOOSE as well. There are a lot more for GOOSE-STRUT (goose-Gus, booze-buzz, etc.)
Travis   Mon May 21, 2007 5:14 am GMT
The main thing, to me, is that there are minimal pairs, albeit few, on /T/ versus /D/ (in the form of noun-verb pairs) and that the two are rather uncomplicated with respect to variation between dialects (with only some relatively minor variation, such as /T/ versus /D/ for "with", with /T/ being more conservative), and the orthographic representation of such would be straightforwards ("th" versus "dh" or "" versus "") so I see no reason to omit the distinction despite such not having all that *many* minimal pairs. This is unlike some things, such as the lot-cloth distinction, which is not really consistent from dialect to dialect in the dialects in which it occurs and which whose representation would likely overly complicate the orthography in question.