Introduction
(Continued from previous page. Page 1 | 2 | 3 | 4 )
Orthography and Transliteration Theories
Introduction
Like other logosyllabic systems, such as Egyptian Hieroglyphic and Japanese, EpM was capable of recording the nuances of spoken language. However, like all writing systems, including our own alphabetic script, EpM writing did not always record or precisely translate speech into graphic symbolic form. For instance, the Latin based script, as used by English speakers, does not have a one to one correspondence between sound and sign. Whereas, Hangul, the script employed in Korea, has a sign sound correlation that represents the Korean language almost perfectly, to the extant that it is claimed that the form of each sign emulated the shape of the tongue when making the sound that the sign represents. Other scripts systematically underwrite certain components of their language because these are more or less obvious to the native speaker/reader. Egyptian, Arabic, and Hebrew scripts all, for example, traditionally omitted vowels or indicated them only through diacritical marks.
EpM likewise underwrote features of spoken language that would have been provided by the native reader or reciter. The transcription of EpM text therefore presumes that we may, with some fidelity, uncover the ancient utterance. Needless to say, the difficulty of this operation makes glyphic transcription one of the most hotly debated and rapidly changing areas in the field today. The position taken here is that while at one level it is very satisfying to be able to recite an inscriptional passage and thereby enliven it, there are problems with relying on this level of transcription for the analysis of glyphic texts. For this reason the first level of transcription is included in this notebook in bold letters. Logograms are written in UPPER case and syllabograms are written in lowercase letters (see below). The second level of transcription takes this information and renders it in a form that attempts to approach spoken EpM. These transcription theories are much debated because differences in renderings produce different translations of words or passages and grammatical interpretations. Without a precise understanding of the form of the language, many grammatical questions are difficult if not impossible to answer. The following section reviews the most recent and promising of the theories. Ultimately, we will find that there is probably more flexibility and ambiguity in the script than any one of these positions currently allow. The fact that EpM was written for almost two thousand years will explain many of the differences between sites and various spellings.
The transcription of texts has until recently not been possible. Before the 1990s, glyphic texts were known mainly through paraphrased translations, which only attempted to transcribe the most transparent spellings. This situation radically changed with the development of a more complete syllabary and logograph table, and a greater understanding of grammatical structures. The questions now surrounding the script concern the presence of complex vowels, polyvalent signs, and the meaning of the texts themselves.
Transliteration is a four-step process. The first step requires the transcription of all the information found in the text. Syllables are written in lowercase while logographs are written in uppercase. Signs within the same glyph block are joined with a hyphen and a question mark, or Thompson numbers, mark unknown signs. The second step takes this information and puts it into a form that more closely resembles the form of the actual utterance. Words are separated by function so that what might in the inscriptions be written in two different glyph blocks would be joined into a single word or the opposite might be true. The systems of lower and uppercase distinction are not used but the passage is written in italics. From this point a third step may be taken where the passage is translated and grammatically analyzed. The final fourth level is the true translation of the passage that renders it intelligible to any educated reader. This fourth step is only rarely preformed.
As can be seen, transliteration at one level requires only a basic knowledge of the phonograms and several dozen logograms. However, there are numerous linguistic issues that make this process more complex. Some of the most important theories about the writing system and orthographic principals are synharmony, disharmony, morphosyllables, vowel inversion or insertion, and historic-pragmatic spelling rules.
Harmonic Theories:
Synharmony: The use of two CV signs to write a CVC word where the second syllable echoes the vowel of the root. This principle was first developed and used by the Russian epigrapher Yuri Knorozov in what were among the first successful phonetic readings of the script.
CV-CV1 = CVC
k’a-ma or K’AM-ma, for k’am
Disharmony: The use of a final CV sign where the vowel of the final CV sign does not repeat the root vowel (V1 ≠ V2) in order to indicate a complex root vowel. In this case, the vowel of the second syllable indicates the nature of the vowel of the first (Wichmann and Lacadena in Notebook 2001, 2004).
Harmony Rule 1
1a CVC-CV1/CV-CV1 > CVC
or CVhC
Harmony Rule 2
2a CVC-Ci/CV-Ci > CVVC,
where V = e, a, u, o
2b CVC-Ca/CV-Ca > CVVC,
where V = i
Harmony Rule 3
3a CVC-Cu/CV-Cu > CV’(V)C,
where V = i, a
3b CVC-Ca/CV-Ca > CV’(V)C,
where V = e, u, o
Example:
ka-b’u-la ka-b’u-la 3-ka-b’u-la
kab’ul kab’ul ux kab’ul
our beans, our beans, our three bags of beans.
Predictive –VL Spelling Rules
Kaufman and Justeson suggested at the 2002 Maya Meetings in Austin that it was likely that the unpronounced vowel of the final CV syllable in spellings is “the one (or a one) in a -V:l suffix that was characteristically suffixed to that root in the Epigraphic Mayan language, which should be best reflected in Greater Tzeltalan languages today” (PMED 31). Kaufman and Justeson offer the most complete outline of this theory, but they are still in the process of collecting data for statistical tests.
Erik Boot (n.d.) has recently presented his own elaboration of a theory countering Lacadena and Wichmann in which he suggests that an ending of Ci or Ca signaled morphemic boundaries and word endings. He further argues that vowel complexity is not normally indicated in the script, except in cases of hypercorrection such as NAH-wa-ja and NAH-ka-na-ja, for nahwaj (was adorned) and nahknaj (was raised; installed). Like Kaufman and Justeson, he contends that the final V of a word final syllable anticipates the most common affixing suffix. Thus in his view, vowel complexity is either not represented or not present in EpM.
Grammatical suffixes: Morphosyllables or Vowel Inversion
Morphosyllables
Morphosyllables are a class of word signs used to indicate verbal and nominal inflection. As signs they stand in between logograms and syllabograms and are identical in form to their syllabic counterparts except that the vowel of the CV syllable is transposed so that the morphosyllable is read a VC syllable, where the value of V is context dependent. The existence of MS is still open to debate (Houston et al. 200).
Morphosyllabic spelling:
Root: k’/ch’am (take)
U-K’AM-V1W, for uk’amaw (he/she/it took
it)
K’AM-AJ, for k’a[h]maj (it was/is taken,
it was/is grasped)
U-cho-ko-V1W, for uchokow
cho-ka-AJ, for cho[h]kaj
Nominal Suffixes:
Root: tun (stone)
U-TUN-ni-VL, for tunil stone
Notice in the Morphosyllabic system certain syllables (ja, wa, u, etc.) are treated as logographic signs, which Houtson et al. suggest were originally logographic in nature and then were later generalized into syllabic signs that could be used in any context. This developmental sequence seems critical to their argument. However, evidence from the development of similar grammatical signs in Japanese, (such as the topicalizer は wa, the direct object marker を wo~o, and the directional へ he) suggests just the opposite situation, as does internal evidence form the Maya script. For these reasons too lengthy to examine in detail here, I will follow the more traditional view that the vowel is simple supplied by the reader, while noting the fact that the Maya scribe early on in the script’s history restricted the range of signs that could spell particular suffixes.
Vowel insertion/prediction
Another equally compelling explanation of vowel transposition argues that the second or thematic vowel of root verbs is under-written and is to be supplied by the reader. In these cases if the word is spelled completely phonetically, the vowel of the penultimate syllable predicts the V to be inserted.
Vowel Inversion spelling:
u-K’AM-wa, for uk’amaw (he/she/it
took/takes it)
K’AM-ja, for k’a[h]maj (it was/is
taken)
u-cho-ko-wa, for uchokow (he/she scattered/scatters
it)
cho-ka-ja, for cho[h]kaj (it was/is scattered)
Nominal Suffixes:
Root: tun (stone)
u-TUN-ni-li, for tunil stone
Historic-Pragmatic Spelling Rules
Historical pragmatic spelling rules, again lacking a definitive label,
see spelling, especially of verbal suffixes, as a systematic complex distinct
from the rules employed in the spelling of nominal roots. A pragmatic
spelling theory sees the regular spelling of verb-final syllables in either
Ca or Ci as indicative of an ancient attempt to systematize and regularize
the spelling of suffixes and provide visually consistency to the script.
This theory does not necessarily exclude other spelling theories but rather
suggests that different sets of rules could inform different paradigms
for historical reasons. This is also the system that will underlie my
organization of verbal suffixes.
For example, Barbara Macleod and others have observed that the wa syllable was probably chosen to spell the transitive thematic suffix because at the time of the script’s development the most common ending of transitive verbs in the declarative was –a, a pattern still preserved in K’iche’, Tojolab’al, and some other languages. In these languages, –a is the most common vowel ending of transitive verbs. Variation occurs when this final vowel mirrors that of the root. Over time this harmonic rule seems to have produced a series of different vowel endings for the transitive verb, -i, -e, -a, -u in Ch’orti’ for example, and -aw, -ow, -uw, -iw, and -ew in EpM. Even during the Classic period an -a(w) thematic was still the most common ending of transitive verbs probably simply for the fact that a is the most common root vowel of CVC transitives.
Choosing a transcription system is difficult. Throughout this current work, I use a system that does not recognize harmonic theories and therefore I do not reconstruct long, rearticulated, or aspirated vowels. I choose this system over others because I have reservations about the ability of users of a syllabic system to analysis individual phonemes. Additionally, and more importantly, I find that harmonic systems do not accurately represent the expected and logical form of verbal suffixes. If one aspect of the paradigm is incorrect, it suggests that the entire system might be in error or that different rules govern different subsystems within the writing system. Because of this uncertainty, I present the actual glyphic spelling in bold and would recommend that one learn to work at this level of transliteration since it most accurately presents the inscriptions.