Loading ...
Sorry, an error occurred while loading the content.

Categorizing English tengwar modes

Expand Messages
  • Dave
    Greetings everyone, at the risk of making a fool of myself again, I ve finally decided to try and organize all known English tengwar samples (excl.
    Message 1 of 6 , Jan 3, 2006
    • 0 Attachment
      Greetings everyone,

      at the risk of making a fool of myself again, I've finally decided to try and organize all known English tengwar samples (excl. Anglo-Saxon/Old English) by Tolkien's own hand into the four categories Orthographic Tehta Modes (OTM), Phonemic Tehta Modes (PTM), Orthographic Full Modes (OFM) and Phonemic Full Modes (PFM). Most of it seemed pretty straightforward, but some of it didn't. Hope I didn't miss anything.

      So, here goes:

      OTM: DTS 5, 10 (second greeting), 11, 62.

      PTM: DTS 39, 41 (partly), 47, 56, 58.

      OFM: DTS 10 (third greeting), 13, 25, 45, 48, 49, 54.

      PFM: DTS 1, 14, 16, 17, 18, 22, 23, 24, 28, 53, 60.

      Doubtful: DTS 15 (really OE/PFM), 36 (probably PTM), 37 (probably PTM).

      Some remarks:

      1) DTS 47 (d'Ardenne Dedication): I agree with M. Bjoerkman that this should be a sample for phonemic spelling ("rEturn" and "kIng" spelled the same [I-tehta] because of identical pronunciation in spite of different English spelling). I'm a bit confused, however, by the dot under the oore in "return". I don't think this is attested in any of the other PTM specimens (alas there aren't many). DTS 41 has what would be the same sound in "i(n)cuRsu", provided it's pronounced in modern English, and here there's no underdot (then again, maybe this part of DTS 41 is to be read according to more classical Latin pronunciation--this is what the rest of the phrase would imply with identical tehta spelling in tengwar for identical letter spelling in Latin, where vowels are pronounced much more uniformly anyway--so that it's harder to discern between phonetic/orthographic spelling).
      The sound occurs nowhere else in other PTM samples.

      2) DTS 36/37: Obviously tehta, but phonemic or orthographic? Against orthographic might speak that in other OTM samples (well, just DTS 5 really), the shorthand for "the" is spelled without an underdot (on the other hand, in DTS 5 there is an underdot under the shorthand for "and", which is not seen under the same shorthand in DTS 10 (second greeting)--leading me to ask: what is the significance of the existence/lack of the dot under these shorthands, and why do both spellings occur within what's basically the same mode).

      I'm also unsure about the spellings of "lie", on which much would hinge. First off, I'm not sure I _what_ we're really dealing with: the first letter is clearly lambe, but then? in DTS 36, it would sem to be followed by two dots on vilya--or is it actually anna, but the flourish from the lambe makes it look otherwise? In DTS 37, it seems more or less clearly to be anna, but strangely this time with three dots on top. Should we really interpret this as "lay", as Chris McKay has done? If not, then the puzzling question remains why the same word would be spelled differently in the two drafts, otherwise clearly in the same "mode". Maybe just a mistake?

      The combinations "ie" and (and potentially "ay") don't occur in the other OTM samples, so we can't make comparisons. Based on for example DTS 62 (and the spelling there of "MichAEl"), though, I'd probably rather have expected something along the lines of I-tehta on yanta for "ie" as orthographic notation.
      Neither does the _sound_ of "lIE" appear in any of the PTM samples. (In the Bombadil/Treebeard modes, this sound is spelled with two dots on vilya/two dots on annay respectively.)

      Interestingly, in OFM samples such as The King's Letter the letter "a" is spelled with vilya, so maybe in DTS 36 we see "a" plus two dots to indicate following -i/-y? This would be bottom-top spelling, though, against the rest of the sample, and, more importantly (since reversed reading direction does occur in diphthongs) also against DTS 37, where, no matter if we read the last word "lie" or "lay", we are clearly dealing with top-bottom spelling, since the A-tehta certainly represents either the sound or the letter of "a".

      _If_ DTS 37 really has "lAY" (which I tend not to believe--why would this famous line from the Ring poem be changed here?), this would probably speak against a PTM spelling, since the same sound is spelled with an E-tehta on anna in DTS 39 (and with a a double dot on a stemless calms [or small Latin "c"] in what Wust has labeled the "Bombadil Mode", but that's of course a full mode...).

      The spelling of "Mordor" (O-tehta for both "o") would seem to be in favour of orthographic spelling, since in modern English pronunciation, the two "o" would almost certainly not be pronounced the same (and should thus be spelled differently in a phonetic mode), but one could of course argue that this is really based on Sindarin pronunciation (and spelling), since the word's taken from that language.

      The appearance of hwesta sindarinwa for the sound in "WHere" is also inconclusive: orthographic to discern this from the simple "w" in "war" or "Westmarch" (DTS 5)? One could also argue, though, that in some variants of English, "WHere" is actually pronounced differently (i.e. with an audible "h"), and then this could also support the theory that DTS 36/37 are phonemic modes.

      Finally, it's interesting that in DTS 5, the silent "e" in "herEin" is marked by an underdot (under roomen, used here since followed by a vowel I think), but no such dot is seen under oore ("wherE") in DTS 36/37.

      Considering all the above, I'd still lean towards calling these two specimens PTM. Maybe I've overlooked something?

      3) Obviously full mode. M. Bjoerkman lists this as an "Old English sample" (I don't know much about OE, but this should be "stone fortress", with "borg" probably a variant of "burg" as in Mundburg [name for Minas Tirith in Rohan] or Hornburg). If we accept this, it should strictly speaking not be listed here, but Wust, for example, also discusses the Steinborg drawing in his article on Tolkien's Phonetic _English_ modes. Maybe because in this case, the difference between old/modern English pronunciation wouldn't be so big? I'm not certain.
      Anyway, the use of vilya for "o" dovetails with the Treebeard page, as Wust points out, but I don't think two dots on stemless calma is seen anywhere else for the sound of "stEIn"? (the Bombadil modes uses this for the sound "wAY", but that's different). Still, the fact that Tolkien wrote a vowel chart (though crossed out) on the upper right of the "Steinborg" spelling, and that these showed phonetic representations for vowels, would also strongly suggest that this is a phonetic mode.

      Hope there aren't too many mixups this time...

      Hisilome




      [Non-text portions of this message have been removed]
    • j_mach_wust
      Dave Hisilome wrote: ... DTS 41 may or may not be phonemic English, but part of it certainly has English phonemics. ... I d say DTS 14 is also orthographic,
      Message 2 of 6 , Jan 3, 2006
      • 0 Attachment
        Dave "Hisilome" wrote:
        ...
        > at the risk of making a fool of myself again, I've finally decided
        > to try and organize all known English tengwar samples (excl.
        > Anglo-Saxon/Old English) by Tolkien's own hand into the four
        > categories Orthographic Tehta Modes (OTM), Phonemic Tehta Modes
        > (PTM), Orthographic Full Modes (OFM) and Phonemic Full Modes (PFM).
        > Most of it seemed pretty straightforward, but some of it didn't.
        > Hope I didn't miss anything.
        >
        > So, here goes:
        >
        > OTM: DTS 5, 10 (second greeting), 11, 62.
        >
        > PTM: DTS 39, 41 (partly), 47, 56, 58.

        DTS 41 may or may not be phonemic English, but part of it certainly
        has English phonemics.

        > OFM: DTS 10 (third greeting), 13, 25, 45, 48, 49, 54.

        I'd say DTS 14 is also orthographic, but DTS 25 is phonemic.

        > PFM: DTS 1, 14, 16, 17, 18, 22, 23, 24, 28, 53, 60.
        >
        > Doubtful: DTS 15 (really OE/PFM), 36 (probably PTM), 37 (probably
        > PTM).
        >
        > Some remarks:
        >
        > 1) DTS 47 (d'Ardenne Dedication): I agree with M. Bjoerkman that
        > this should be a sample for phonemic spelling ("rEturn" and "kIng"
        > spelled the same [I-tehta] because of identical pronunciation in
        > spite of different English spelling). I'm a bit confused, however,
        > by the dot under the oore in "return". I don't think this is
        > attested in any of the other PTM specimens (alas there aren't many).

        This doesn't confuse me, since a representation of the stressed rhotic
        vowel sound of "fur" as schwa + r is also attested in the "Bombadil"
        phonemic "full writing" or in Sarati samples such as R22.

        > 2) DTS 36/37: Obviously tehta, but phonemic or orthographic? Against
        > orthographic might speak that in other OTM samples (well, just DTS 5
        > really), the shorthand for "the" is spelled without an underdot (on
        > the other hand, in DTS 5 there is an underdot under the shorthand
        > for "and", which is not seen under the same shorthand in DTS 10
        > (second greeting)--leading me to ask: what is the significance of
        > the existence/lack of the dot under these shorthands, and why do
        > both spellings occur within what's basically the same mode).

        Another source for the dots below in orthographic tehtar modes is the
        passage of Appendix E that describes DTS 5. It also states that
        there's a dot in unstressed "and", but no dot in the abbreviations of
        "the, of, of the". It's also interesting that the word "of" isn't
        abbreviated in either sample. I don't think that an unusual spelling
        of "the" is very significative, since this word has an unusual
        spelling in virtually any text in Tolkien's alphabets.

        > I'm also unsure about the spellings of "lie", on which much would
        > hinge. First off, I'm not sure I _what_ we're really dealing with:
        > the first letter is clearly lambe, but then? in DTS 36, it would sem
        > to be followed by two dots on vilya--or is it actually anna, but the
        > flourish from the lambe makes it look otherwise?

        I'd say it's a vilya; I think there's no sample of a stroke extending
        to a letter that isn't supposed to have one.

        ...
        > the puzzling question remains why the
        > same word would be spelled differently in the two drafts, otherwise
        > clearly in the same "mode". Maybe just a mistake?

        That's what I think. I can only explain the transcription of DTS 36
        "lie" as a mistake.

        ...
        > Neither does the _sound_ of "lIE" appear in any of the PTM samples.
        > (In the Bombadil/Treebeard modes, this sound is spelled with two
        > dots on vilya/two dots on annay respectively.)

        Based on analogy with the phonemic full modes, we can assume that the
        second part of the /aj/ diphthong will be spelled like the second part
        of the /ej/ diphthong which is attested in DTS 39 "praise": with anna;
        and that its first part will be spelled with the normal a-tehta. So
        we'd expect three dots on anna. This is exactly what we have in DTS
        37, so we can identify it with "lie" /laj/.

        > Interestingly, in OFM samples such as The King's Letter the letter
        > "a" is spelled with vilya, so maybe in DTS 36 we see "a" plus two
        > dots to indicate following -i/-y? This would be bottom-top spelling,
        > though, against the rest of the sample, and, more importantly (since
        > reversed reading direction does occur in diphthongs) also against
        > DTS 37, where, no matter if we read the last word "lie" or "lay", we
        > are clearly dealing with top-bottom spelling, since the A-tehta
        > certainly represents either the sound or the letter of "a".

        This is not only a question of top-bottom vs. bottom-top, but more
        importantly a question of tehtar mode vs. "full writing". The nature
        of most English diphthongs is that the truely vocalic part is the
        first one. It's the part that carries the stress, it's the part that
        can be lengthened more naturally when singing ("joooooooy" is much
        more natural than "joyyyyyyyy", it's the part that allows for more
        variation (the second part must be either /j/, /w/ or, in certain
        dialects, schwa). So if we are to decide which of the two parts is a
        vowel and which part is a consonant, the answer must be that it's the
        first part which is vocalic and the second part which is
        consonantical. In a tehtar mode, the vocalic part must be written with
        a tehta, while the more consonantal part is usually written with a
        tengwa (though sometimes with a tehta as well); vice versa in "full
        writing".


        I guess I'm at least partly responsible for the idea that DTS 37 reads
        "lay" and not "lie" because I was fixed on the idea that it were an
        orthographic mode. I now believe I was wrong. We have two explanations:

        (1) DTS 37 is spelled orthographicly but misspells the ring poem (the
        original says "lie") AND drops the dot below "where" AND uses
        different strange representations for "the", "of".

        (2) DTS 37 is spelled phonemicly.

        Explanation (2) is very much simpler and therefore preferrable. So I
        think now, with Måns, that I was mistaken originally and that DTS 37
        and DTS 36 are both phonemic modes.


        DTS 15:

        > 3) Obviously full mode. M. Bjoerkman lists this as an "Old English
        > sample" (I don't know much about OE, but this should be "stone
        > fortress", with "borg" probably a variant of "burg" as in Mundburg
        > [name for Minas Tirith in Rohan] or Hornburg). If we accept this, it
        > should strictly speaking not be listed here, but Wust, for example,
        > also discusses the Steinborg drawing in his article on Tolkien's
        > Phonetic _English_ modes. Maybe because in this case, the difference
        > between old/modern English pronunciation wouldn't be so big? I'm not
        > certain.

        By now, I consider only the vowel chart to be phonemic modern English.
        It's a complete list of all seven vowels that are found in Tolkien's
        phonemic English transcriptions that distinguish between the "nut"
        vowel and schwa. No other language that I'd know of would require
        these seven vowels, so I think it's safe enough to assume that this is
        phonemic English "full writing". The rest of DTS 15 is basically a
        name (and it's not really Old English either).

        ---------------------------
        j. 'mach' wust
        http://machhezan.tripod.com
        ---------------------------
      • hisilome
        Alright, many thanks to J. Mach Wust for his comments! In brief: I agree that DTS 14 is also _orthographic_ full writing (the vowels in particular exactly
        Message 3 of 6 , Jan 4, 2006
        • 0 Attachment
          Alright, many thanks to J. 'Mach' Wust for his comments!

          In brief:
          I agree that DTS 14 is also _orthographic_ full writing (the vowels in
          particular exactly match the usage in the King's Letter, and also
          those in DTS 13, as one would probably have expected), this was my
          oversight.

          As for DTS 25, I was also wrong. This is a _phonemic_ full mode
          sample--though with some peculiarities, such as the dot under oore in
          "enter", not seen in the Bombadil Mode--unfortunately, this sound
          doesn't occur on the Treebeard page, with which most of the sounds in
          DTS 25 neatly match.
          _Most_ I say, since for example I'd say the sounds in "spEAk" (DTS
          25) and "mEEt" (DTS 24: Treebeard Page) are identical, yet DTS 25 in
          the first instance spells acute accent on stemless calma (at least it
          appears so in the reproduction in HoMe VI--then again, maybe it's
          just a particularly "rounded" short carrier...?), while DTS 24
          represents the sound in various other ways: once, in "mEEt", with a
          short carrier (sorry, after all I just can't see the acute accent on
          top of it [a very faint smudge at most...but that doesn't quite
          convince me. ;)]). The same sound is clearly spelt with a chevron or
          breve on top of a short carrier (the same diacritic that's used for a
          preceding vowel-"y" in English OTM modes) in the word "lEAgue" in DTS
          24, while in "TrEEbeard" we indeed very clearly encounter the acute
          accent on top of a short carrier. Interestingly, the latter spelling
          is also seen in the second occurence of "spEAk" in DTS 25--so maybe
          the first spelling of the word really just features a rather "curvy"
          short carrier...

          The fact remains that DTS 25 belongs in the category PFM, also
          because the vowels are clearly represented by different tengwar than,
          let's say, in the King's Letter or the Mazarbul page.

          DTS 36/37: I'm glad we agree that they should after all be considered
          phonemic, thanks for providing more detailed arguments for this
          conclusion. I concur that to maintain orthographic status for DTS 37,
          one has to make too many concessions/assume to many "mistakes" on
          Tolkien's part. On the other hand, if we just assume one single
          mistake in DTS 36 (misrepresentation in tengwar writing of "lie"),
          then the rest of both samples can clearly be seen as phonemic
          (with "Mordor" spelled according to Sindarin
          convention/pronunciation).

          As for DTS 15, since "Steinborg" is Old Norse, not Old English,
          (though it does mean "stone fortress" after all), I guess it really
          shouldn't be included here at all...

          So the revised list is:

          OTM: DTS 5, 10 (second greeting), 11, 62.

          PTM: DTS 36, 37, 39, 41 (partly), 47, 56, 58.

          OFM: DTS 10 (third greeting), 13, 14, 45, 48, 49, 54.

          PFM: DTS 1, 16, 17, 18, 22, 23, 24, 25, 28, 53, 60.

          Hisilome
        • j_mach_wust
          Dave Hisilome wrote: ... I think these dots below óre can be explained: It s a consequence of óre being used for any r-sound in this sample. Therefore, a
          Message 4 of 6 , Jan 4, 2006
          • 0 Attachment
            Dave "Hisilome" wrote:
            ...
            > As for DTS 25, I was also wrong. This is a _phonemic_ full mode
            > sample--though with some peculiarities, such as the dot under oore
            > in "enter", not seen in the Bombadil Mode--unfortunately, this sound
            > doesn't occur on the Treebeard page, with which most of the sounds
            > in DTS 25 neatly match.

            I think these dots below óre can be explained: It's a consequence of
            óre being used for any r-sound in this sample. Therefore, a syllabic
            r-sound needs a dot below just like for instance a syllabic m-sound:
            It distinguishes the syllabic r/m-sound from the plain r/m-sound. If
            both rómen and óre are used for r-sounds, then that distinctive dot is
            not required because there already is a distinction.

            A mode with one r-letter only would be ideal for the transcription of
            rhotic varieties of English and in any case easier than the widespread
            modes with two r-letters. However, there's a major drawback: It's not
            clear how the sounds of w and wh would be represented in these modes.
            If we assume that all modes where óre is the only r-letter work the
            same, then w would be represented by rómen (as in DTS 10, 13, 50, 51,
            60) and wh by a ligature of halla and rómen (only attested in the Old
            English samples DTS 50, 51); however, we also might assume that the
            modes where óre is the only r-letter do not share a common
            representation of w and wh but that some might use vala and hwesta
            sindarinwa like other modes.


            I believe the "chevron" in DTS 24 is a mistake: It was begun a single
            dot and then amended to an acute. The use of the acute is clearly
            recurring in that sample as an indication of length, so I think
            there's not much doubt.


            As for DTS 36/37: How do you pronounce Mordor in English? I guess you
            rhyme it with "border". I rather pronounce it as "more door" (with the
            stress on "more"), so to me, Tolkien's transcription fits very well
            within a phonemic English mode.


            > As for DTS 15, since "Steinborg" is Old Norse, not Old English,
            > (though it does mean "stone fortress" after all), I guess it really
            > shouldn't be included here at all...

            I disagree. For sure, the "Steinborg" part shouldn't be included, but
            the phonemic English vowels should.

            ---------------------------
            j. 'mach' wust
            http://machhezan.tripod.com
            ---------------------------
          • Dave
            ... I think these dots below óre can be explained: It s a consequence of óre being used for any r-sound in this sample. Therefore, a syllabic r-sound needs a
            Message 5 of 6 , Jan 4, 2006
            • 0 Attachment
              "j_mach_wust" wrote:


              >Dave "Hisilome" wrote:
              >...
              > As for DTS 25, I was also wrong. This is a _phonemic_ full mode
              > sample--though with some peculiarities, such as the dot under oore
              > in "enter", not seen in the Bombadil Mode--unfortunately, this sound
              > doesn't occur on the Treebeard page, with which most of the sounds
              > in DTS 25 neatly match.

              I think these dots below óre can be explained: It's a consequence of
              óre being used for any r-sound in this sample. Therefore, a syllabic
              r-sound needs a dot below just like for instance a syllabic m-sound:
              It distinguishes the syllabic r/m-sound from the plain r/m-sound. If
              both rómen and óre are used for r-sounds, then that distinctive dot is
              not required because there already is a distinction.

              <<< Why, yes. That would make perfect sense! I'll go along with that. >>>


              A mode with one r-letter only would be ideal for the transcription of
              rhotic varieties of English and in any case easier than the widespread
              modes with two r-letters. However, there's a major drawback: It's not
              clear how the sounds of w and wh would be represented in these modes.
              If we assume that all modes where óre is the only r-letter work the
              same, then w would be represented by rómen (as in DTS 10, 13, 50, 51,
              60) and wh by a ligature of halla and rómen (only attested in the Old
              English samples DTS 50, 51);

              <<< And thus maybe not such a good solution for a modern English mode? I
              admit, I also dislike that ligature letter for purely aesthetical reasons...
              >>>

              however, we also might assume that the
              modes where óre is the only r-letter do not share a common
              representation of w and wh but that some might use vala and hwesta
              sindarinwa like other modes.

              <<< You mean, like in DTS 36/37? Too bad that in these samples we cannot
              really be sure how "r" in front of a (pronounced) vowel would be written...

              The Treebeard page also shows hwesta sindarinwa for "wh", and uses vala for
              "w", but it discerns between pronounced and dropped "r" (roomen/oore),
              except in the word "fragment" in "This is a fragment", a phrase added to the
              margin of the page--but this is probably a mistake anyway, considering that
              the "g" in "fragment" is apparently also misspelt (and that in the text
              proper, "fragment" is spelt with roomen, and the "R-rule" generally
              observed).

              Hm. Any other samples that would be relevant? >>>



              I believe the "chevron" in DTS 24 is a mistake: It was begun a single
              dot and then amended to an acute. The use of the acute is clearly
              recurring in that sample as an indication of length, so I think
              there's not much doubt.

              <<< Well, that's of course a possible (and pretty likely) interpretation, it
              just looks like quite a neat chevron to me. I guess you'd also maintain that
              the first occurrence of "spEAk" in DTS 25 has indeed just a very "curvy"
              short carrier with acute accent on top (still _looks_ a lot like a small "c"
              to me). Then again, written, probably rather hastily, on the top margin of
              the page, we may have to make allowances. :)
              As for "mEEt", I still honestly can't see anything on top of the short
              carrier, not even a smudge really. I mean if there is anything, it's so much
              fainter than all the other accents as to be invisible to my eyes (I've only
              got a paperback edition of HoMe VI, but in my experience the black-and-white
              reproductions in the hardcover volumes aren't necessarily much cleareer).
              On the whole, though, of course you're right that we have recurring evidence
              of the acute accent being used as an indicator of length--so even if "meet"
              lacks it, it's probably just a slipup. >>>


              As for DTS 36/37: How do you pronounce Mordor in English? I guess you
              rhyme it with "border". I rather pronounce it as "more door" (with the
              stress on "more"), so to me, Tolkien's transcription fits very well
              within a phonemic English mode.

              <<< Hm hoom. Now that you mention it: yes, I might pronounce it like that
              (rhyming with "border") when I'm talking very quickly, but in slower speech,
              I think I pronounce it pretty much the way you described it. Could be our
              knowledge of Sindarin pronuncation affecting our enunciation, of course. How
              many two-syllable words in English end in "-or"? In longer words (e.g.
              "predator" or "interrogator") the "-or" is clearly but a schwa, I guess
              because it's so far removed from the main stress? Well, this is not my forte
              and getting off-topic anyway. Bottom line, I basically agree with your "more
              door" description--the same "o"-sound twice, just pronounced weaker in the
              second syllable. Makes the case only stronger for English PTM here. >>>


              > As for DTS 15, since "Steinborg" is Old Norse, not Old English,
              > (though it does mean "stone fortress" after all), I guess it really
              > shouldn't be included here at all...

              I disagree. For sure, the "Steinborg" part shouldn't be included, but
              the phonemic English vowels should.

              <<< I won't argue with that. I somehow had the idea in my head that only
              "real text" (and even if it just consists of a name) would count as a
              tengwar sample, but that's of course nonsense, as for example DTS 9 (Tengwar
              table from LotR appendices) shows. >>>

              Greetings,

              Hisilome
            • hisilome
              Just putting a small matter to rest. ... Turns out I was wrong here: Having had a look at the hardcover edition (in this case, the Houghton Mifflin one) of
              Message 6 of 6 , Jan 9, 2006
              • 0 Attachment
                Just putting a small matter to rest.

                In an earlier post, I wrote:

                >As for "mEEts" [in DTS 24], I still honestly can't see anything on
                >top of the short carrier, not even a smudge really. I mean if there
                >is anything, it's so much fainter than all the other accents as to be
                >invisible to my eyes (I've only got a paperback edition of HoMe VI,
                >but in my experience the black-and-white reproductions in the
                >hardcover volumes aren't necessarily much clearer).

                Turns out I was wrong here: Having had a look at the hardcover
                edition (in this case, the Houghton Mifflin one) of HoMe IV, I have
                to say that there _is_ a distint actue accent on "mEEts"--faint, but
                not very much more so than some of the others in the same sample.

                Rather "disturbingly", though, I also discovered another acute accent
                that hadn't caught my attention before. It's similarly faint, but
                equally distinct, on the vilya in "lOst". I was under the impression
                that that sound was actually short (especially when compared
                to "mEEts" or "decEIved")--so why would it receive an "andaith"? Then
                again, this is probably just a matter of variants of pronunciation...

                While I was at it, I also had another look (in the hardcover edition)
                at the first occurence (top margin of manuscript page) of "spEAk" in
                DTS 25. In this much clearer reproduction, the tengwa representing
                the "EA"-sound still looks a lot like a stemless calma, especially
                when compared to the other examples for a short carrier occurring in
                this specimen. And the acute accent on top actually reminds me a bit
                of a circumflex (with the left stroke markedly longer than the
                right). As I speculated before, maybe Tolkien was writing rather
                hastily here...
                So we could be dealing with a "reversed" incidence of what we saw in
                the word "lEAgue" in DTS 24: there J. 'Mach' Wust thinks that Tolkien
                emended what began as a single dot into an acute accent, thus
                producing a breve or "chevron"--in "spEAk", we might then be dealing
                with the same situation, only that in this case the emendation turned
                the dot into what looks somewhat like a circumflex. Interestingly,
                though, all the actual dots (both single/double, under/over) in this
                sample look rather like circles, not really dots at all. ;)

                Hisilome
              Your message has been successfully submitted and would be delivered to recipients shortly.