Literacy among the Pre-Columbian Maya: A Comparative Perspective

~ Stephen Houston lI iteracy, the ability to link language and script, forms one of the most important topics in twentieth-century (e.g., Ong 1982). On a practical ters of pedagogy 1972;Street 1984), and on a theoretical ken and written level literacy relates to mat- and societal development (Cipolla an understanding level to the interaction in the Pre-Columbian among the Maya of Mesoamerica, for literacy world, and complexity. Maya: A particularly This essay explores against a backdrop formation from other parts of Mesoamerica of comparative and the Old World. Comparative in- Its aim: on literacy that goes beyond the particulars of Maya culture, yet enriches the debate on ancient literacy and its consequences with evidence from a region that has been persistently neglected or misunderstood by writers on the subject (e.g., Goody 1987: 22, 23). Perspectives on Ancient Literacy The subject of ancient universal definition literacy writing-that presupposes of literacy, variable and culturally several questions: or should literacy determined Is there be defined ways? How does reading depart from to be studied at leisure in a setting divorced nature do certain writing those of a logosyllabic kind, inhibit difficult modern inadequate sampling. literacy in the to answer is the nature of the of we cannot study ancient literacy terms, this stems from the problem of Often, the relative tions reveals little or nothing read and write linguists, who have the benefit living speakers to survey and interview, by direct means. In archaeological of systems, particularly the growth of widespread and reflective mindsets? 1 And how many could ancient world? Unlike message, from the oral recitation that message? By their What makes these questions a in highly is, to what extent does a text record a complete available evidence. lack or abundance about literacy, of inscrip- about how many people could write, who could write, and to what degree. Other forms of writing simply failed to survive, so that negative evidence cannot be interpreted as a sign of limited literacy Pre-Columbian where scribal arts reached an extraor- dinary degree of accomplishment to present a perspective of spo- change (Chafe 1987: 263). Still undeveloped, however, is of literacy the Maya evidence among the 1969; Sanderson language and the intensity of cognitive and Tannen 1987; Goody Literacy linguistics (Harvey 1966: 586, 590; Mann 1985: 206; Johnston 1983: 66). We can escape from this impasse by dealing with the few direct clues that survive and by examining tions derived from the study of comparable indirect clues-expecta- writing systems, from what Perspective the ancients themselves said about literacy, and from an understanding graphically by imagining a t of how scripts functioned production on the x axis and n is charted from left to right, al in society. With such evidence we can addressthe first question: What are the defi- and upper right corners repre of scribal accomplishment; ~ standard of literacy, but lie wi nitions of literacy? The more abundant data from the Old World reveal an enormous range of meanings (Schofield 1968: 313-314; Thomas 1989: fluent reading and exegesis,I would fit into the lower right Defining Literacy a Medieval scribe copying a 19-20, 33-34). In the first place, the ability to read was not necessarily the same as the ability to write (Clanchy 1979: 183; Harris 1989: 5), slot in the upper left. There is another point to k since the latter often involved greater preparation and skill. Ironically, in some societies of Late Antiquity, a person could be regarded as "lit- duction and responsetouchl within the scribal tradition a erate" or even be termed a "scribe," yet show minimal competence; that is, some people could write but not read! The scribes from Fayum, son's lifetime (not always pn pertise change within a sing changes occur in responset. Egypt, barely able to scrawl their names or copy another's script, are notorious in this regard (seealso Troll 1990: 113-114); yet witness their anger when characterized as illiterate (Youtie 1971: 248), a situation 1983: 583). Ac single lifetime can be illustr tion (Baines that compelled the Emperor Justinian to avoid confusion in his notarial system by recognizing various degrees of literacy (Youtie 1971: 254, tinuum, yielding a distincti, those of other individuals. ( 261). At the other extreme lies the Chinese conception of "full literacy" (Rawski1979: 4-5). In imperial China, literacy presupposed a thorough grounding in Confucian learning. Nonetheless, more objective measures tradition would also presen for the purposes of qualitat question of defining literac (literate vs. illiterate) nor in stantial numbers of peopll of literacy in China signal the widespread ability to use Chinese characters as early as the fifth century A.D., although exegetical skill is likely to have been limited in all periods (Rawski 1979: 5). material (Boring 1979: 1).4 The essential point here is that literacy was often defined in cultural terms and in ways that make it difficult to evaluate the meaning of ancient references to the numbers of people who could read or write.2 For this of reading and writing, and but by the standards of ind reason alone, we should adopt a flexible definition of literacy. Attempts to define "full" literacy in rigid terms, such as Eric Havelock's emphasis Orality and Literacy: The I on a personalized literary tradition recorded in alphabetic script (e.g., Havelock 1982: 6, 27, 57), cast the term in ways that seemtendentious, Another vital question co mance leading to invidious comparisons with Classical Greece. By liberating the word from Havelock's Classical grip, we can acknowledge not so and writing. In th us since Milman Parry fir: much different kinds of literacy, as a continuum of scribal practice that ranges along two axes. poetry (Parry 1971). Acco! the Iliad and the Odysse The first axis is production ("writing"), which extends from the Fayum scribe's crude signature to the calligraphic masterworks of Chinese man- poetic allusion. In other' corded in alphabetic scri~ the repetitions, the metri, darins. The second axis charts the response to an encoded message ("reading"), ranging from the bare perception of meaning or sound to a facility with detailed exegesis. States of literacy may be illustrated Stephen Houston 28 poet composing for an al More recently, Eric Ha ..'J graphically by imagining a two-dimensional coordinate system, with production on the x axis and response on the y. If ascendin~competence is charted from left to right, and from top to bottom, then the lower left and upper right corners represent, respectively, of scribal accomplishment; standardof literacy, but lie within the tradition fluent reading and exegesis, but incapable would fit into the lower right corner; a Medieval scribe copying the nadir and pinnacle the terms of reference are not an absolute itself. A person capable of of effective draughtmanship, a craftsman carving a Maya text or a passage composed by another would fit a slot in the upper left. There is another point to keep in mind. The biaxial continuum of pro- duction and response touches on literacy at both the personal level and within the scribal tradition as a whole. Just as ability shifts within son's lifetime (not always progressively!), pertise change within so does the level of scribal ex- a single society. In the case of ancient Egypt, such changesoccur in response to moments of political tion (Baines 1983: 583). Accordingly, single lifetime can be illustrated tinuum, yielding a distinctive thoseof other individuals. and economic the trajectory by a series of positions within "signature" transi- of literacy within a the con- that might be compared with On a broader scale, changes within the scribal tradition would also present a distinctive for the purposes of qualitative question of defining a per- literacy. form.3 By employing comparison, An acceptable (literate vs. illiterate) nor inflexible, the chart we have also addressed the definition is neither binary such as a definition stantial numbers of people engaged in the habitual insisting on subreading of literary material(Boring 1979: 1).4 Rather, literacy varies along the dimensions of reading and writing, and in ways delimited but by the standards of individual Orality not by universal standards, scribal traditions. and Literacy:The Limits of Writing Another vital question mance and writing. us since Milman concerns the relationship In theoretical oral perfor- has been with the oral, bardic basis of Homeric to Parry, Homer "improvised," the Iliad and the Odyssey by following poetic allusion. In other words, although strict guidelines not wrote, of meter and the epics were eventually corded in alphabetic script, their composition the repetitions, the metrical between terms, this question Parry first discerned poetry (Parry 1971). According 'II originated re- in performance; rigor, assisted the memory of a nonliterate poet composing for an audience More recently, Eric Havelock in Dark Age Greece. has reemphasized the oratorical Literacy nature 29 among the Maya . edge depended ultimately on I (Chafe and Tannen 1987: 397; written texts were not so much tl of earlywriting (Havelock1982, 1986), although with a recognition that the Homeric epics represent not pure oral composition, but an attempted reconciliation between spoken poetry and a novel system of writing. To Havelock, literacy was an "act of interpenetration of letters into an messages to be fleshed out by I bodied "recitation literacy" (Havelock 1982: 5): reading was not silent, 21). In practical terms, this me reader of an early Sumerian doe and a tradition of oral discourse, of reciting from memory and of public performance, still represented the principal means of aesthetic and but tacit information (Civil and E holds true for Buddhist scriptur intellectual satisfaction(Balogh1927; Harvey 1966: 588; Knox 1968: 435; Cartledge 1978: 28; Robb 1978: 32; Stoddart and Whitley 1988: 762-763).6 tion. By careful attention to pel the full spiritual benefits of the The conclusion that recitatiol Other evidence suggeststhe prevalence of recitation literacy in the Classical and pre-Modern world, regardless of the script being used. The earliest Greek writing is likely to have functioned at first as a mne- existence of "primary orality" to cultures untouched by kno monic device Oohnston 1983: 67), a means of stabilizing memory. But 1989: 1). Cultures of primary 01 at the same time feats of memorization continued to be highly prized ior and by speech or other SOL no society with writing can e) from the information assemble oral situation (Havelock 1982: 45).5 Even early alphabetic script em- ing recallsthe work of Walterj (Harris 1989: 30-32). In fact, Plato and other authors noted with regret that writing brought "not improved memory but forgetfulness, by providing the literate with an external device to rely on" (Harris 1989: 30; see also Ong 1982: 15; Thomas 1989: 32-33). A similar preference for the spoken word occurs in medieval Europe, where documents were have continued to operate at d much of the Near Eastand the 1989: 283-286). The key diffe removed by incremental degre rize. Eventually, such changE often regarded suspiciously (oral witness being in most casespreferable), and dictation still served as the primary vehicle for literary composition (Clanchy 1979: 211,219; Doane 1991: xii-xiii). Dictation and vocal to oral discourse (Havelock 1 reading had the added merit of permitting the nonliterate to participate (Clanchy 1979: 219). extreme suggested by Ong (19 Recitation literacy has other implications as well, the most important being the partial nature of written communication. To many, the chief significance of written texts is that they are compact, canonical, Writing and Technological DE and authoritative (Harris 1989: 39). Direct refutation of their meaning is impossible, for, after "absolutely total and devastating refutation, [a 1982), as well as that of Goe written text] saysexactly the same thing as before" (Ong 1982: 79). Yet ary impact and cognitive con~ Havelock, who, unlike Gooe One of the more controversii 55-56; this understanding places insufficient emphasis on the interplay between writing, reading, and exegesis (as well as being in itself an ill-founded as expand his earlier views" unique in its sophistication an "wealth of detail," "depth of I and even analytical abstractic notion, for, as an anonymous reviewer points out, "Texts, all by themselves, don't say anything at all"). Often, recorded texts serveas points of departure for performances or further elaborations of their message.They do not stand alone, but, rather, must be read by someone with a com- 24) that had not been recorde through its efficient recordin: prehension of the context and broader meaning, by someone who will take cues from the script. For example, alongside Sumerian cuneiform, there apparently existed a strong oral tradition in which "literate knowl- see also Ong 1982: Stephen Houston 30 ability to relieve the burdens I result: a form of writing capa i' edgedependedultimately on oral reformulations of that knowledge (ChafeandTannen 1987: 397; Heath 1986; Civil 1972). Accordingly, writtentextswere not so much transliterations of utterances asschematic messages to be fleshed out by recitation and performance (Civil 1972: 21).In practical terms, this meant that, to understand a text fully, the readerof anearly Sumerian document had to be familiar with unwritten buttacitinformation(Civil and Biggs 1966; Larsen1988: 187). The same holdstruefor Buddhist scripture, which also depends on oral elaboration.Bycareful attention to performance and exegesis,the reader gains thefull spiritual benefits of the text (Scollon and Scollon 1981: 45). Theconclusionthat recitation literacy characterizes much early writingrecallsthe work of Walter Ong (e.g., 1982), who has proposed the existenceof "primary orality" (Ong 1982: 11). This concept pertains to culturesuntouched by knowledge of writing or print (McKitterick 1989:1).Culturesof primary orality transmit messagesby bodily behaviorandby speech or other sounds (Havelock 1986: 65). By definition, nosocietywith writing can exist within this framework. Yet, to judge fromthe information assembled above, at least part of the mindset may havecontinued to operate at different times and varying intensity within muchof the Near Eastand the Classical world (Ong 1982: 11; Thomas 1989:283-286). The key difference was the presence of writing, which removedby incremental degrees at least some of the pressuresto memorize.Eventually, such changes would have resulted in modifications to oral discourse (Havelock 1986: 101), although probably not to the extremesuggestedby Ong (1982: 59). Writingand Technological Determinism Oneof the more controversial aspects of Havelock's work (Havelock 1982),as well as that of Goody and Watt (1963; Goody 1977, 1987: 55-56; see also Ong 1982: 78), is their insistence on the revolutionaryimpact and cognitive consequences of certain scripts. According to Havelock, who, unlike Goody, failed "to modify and qualify as well asexpand his earlier views" (Halverson 1992: 301), the alphabet was uniquein its sophistication and potential. For the first time it permitted a "wealth of detail," "depth of psychological feeling," personal reflection, andeven analytical abstraction (Havelock 1986: 11; Logan 1986: 2024)that had not been recorded before? The alphabet made this possible through its efficient recording of sounds, its lack of ambiguity, and its ability to relieve the burdens of memorization (Havelock 1982: 61). The result:a form of writing capable of divorcing itself from oral discourse, Literacy 31 among the Maya r ... ~ I. ~ from public as opposed to internal, private thought-in system that created "thinkers" cognitive out of "bards" impact of the script would oral performance 1982: 11). The be felt gradually, as the link with became more and more tenuous. In Havelock's ion, the mere fact that writing "literate"; short, a writing (Havelock opin- was present did not mean the society was rather, literacy was a state of mind enabled by a new kind of scribal technology (Havelock 1982: 57; Chafe and Tannen 1987: 392). A second result of the alphabet was that writing since its simplicity permitted became "democratized," ready adoption by an ever-larger circle of people (Havelock 1982: 83; Goody 1987: 55; Cross 1989: 77-78). Havelock believed syllabaries, inhibited the reverse was also true. Other scripts, especially change, or at least modulated ways. Their very form, their lack of efficient analysis of sound, made the scripts unwieldy by non-elites (Cross 1989: quired a disciplined memorization can recognize of complex syllabic writing inevitably it was impossible paraphrased into easily recognizable and led to a "cautious 1986: 8-9). is much been influenced ment of writing as the pinnacle earlier to criticize 75,96; the inherent deficiencies of here. In the first place, Havelock scheme achievement, a triumph has clearly for the develop- views the alphabet made possible by in the Near East. When he and Gelb discuss other, non-Western either the derivative 1982: into an alphabet 72). systems (1963). Like Gelb, Havelock of scribal it of vocabulary, (Havelock by Ignace Gelb's evolutionary innovations especially 1982: to orality; statements or narratives an economy of sentiment" for they blur and conceal script (Havelock There forms, encouraged restriction 1982: 53). With This is why translations of syllabic originals are misleading, syllabic simplified re- 1982: the field of expertise (Havelock to escape an attachment oral originals, to be used proficiency shapes (Havelock "narrowed and use the script" and inferior and unlikely 77). In Chinese, for example, 52), and the emphasis on calligraphy which thought in restrictive communication scripts, they do so with the aim of demonstrating nature of these writing relative to Near Eastern and Mediterranean systems or their weaknesses writing (e.g., Gelb 1963: 59). The result is often misguided and ethnocentric. For example, Havelock takes the position that, for the Japanese, "purely oral habits of thought and experience. . . have survived more tenaciously in . . . Japan" (Havelock 1982: 348) and that "the free production own script will remain difficult" (Havelock of novel statement in (their) 1982: 347). Since Havelock fails to provide any citations to support these remarks, we must assume that he is revealing more about his preconceptions iarity with the Oriental evidence. than about his famil- In any case, the cultural achievements Stephen Houston 32 of Chineseand Japanesecivilizations make it difficult to accept the viewthat Oriental scripts imposed insuperable limitations on analytic thoughtand philosophical reflection (Gough 1968: 83-84; DeFrancis 1989:244-47). Remember: it was a logosyllabic script that recorded theHeianTaleof the Cenji, an account full of exquisite dissections of emotionaland aestheticstates(Morris 1964: 183-210).8 Otherevidence further undermines the hypothesis of technological determinism.Usually it is society, not script, that determines who can readand in what way. In Heian Japan, women used syllabic kana becausethe prestigiousChinese characters were deemed more suitable for "serious"writing (Morris 1964: 212).9 Despite the inherent difficulties of Chinesescript, there is ample evidence that many people were literate (Rawski1979: 22-23; Sampson 1985: 162), not least because the high prestigeof the writing encouraged people to learn it (Keightley 1989: 192).10 Conversely, thealphabet,which,accordingto Havelock,democratizedliteracy, had effects that were neither uniform nor predictable (Stoddart andWhitley1988:771).TheSpartansusedthe alphabet,but sparingly, sincetheydid not have the same political uses for writing that theAtheniansdid (Cartledge1978: 25; Harvey 1966: 623, 628).Inother words,literacy did not come about because a script compelled it, but becauseauspicioussocial conditions favored the adoption of reading andwriting (Cartledge 1978: 25). Even in areas with the alphabet, the extentof literacy and the rate in which it expanded or declined varied tremendouslythroughtime and space Uohnston 1983: 64; Stoddart and Whitley1988: 771). In all likelihood, these variable patterns were the productof historical and social processes, and not intrinsic to the script itself(Finnegan1988: 158-159). Thereis another problem with Havelock's hypothesis. In characterizinglogosyllabicwriting as cumbersome and inefficient,Havelock misses the point that writing is only one part of an overall system of communication.l1Ifa scriptis read,its messagefleshedout by oraldisputation and elaboration, then it can be a more efficient, more complete transmissionof that message than if the script were to stand alone in silent ambiguity. Havelock himself mentions the continued importance of oral disputationand recitation literacy in Classical Antiquity, yet describes thisnot infunctionalterms, but as a stage of development that preceded full literacy(Havelock 1982: 11). Perhaps a better way of viewing efficiency is less on an absolute scale,with the alphabet as the final achievement, than according to howwriting was used within a particular society (Harris 1989: 324). For example, unlike alphabetic script, which emphasizes differences in lan- Literacy among the Maya 33 I J I guage, Chinese writing lends itself to the unification of a linguistically diverse country, the literate portions of which could read the same logograph in different languages (Goody 1987: 282). A similar pattern occurs in Central Mexican writing, which, as an "open" writingsystem, bridges areas of relative linguistic diversity and cultural homogeneity (Houston 1993b). By alphabetic standards, these scripts are cumbersomebut can it be said that they are inadequate for the purposes defined by their makers? Nor is it clear that nonalphabetic writing, particularly hieroglyphic script, is inherently inefficient in the ways that it conveys meaning. Mayan glyphs contain both phonetic clues and easily recognizable pictorial signs. There is evidence that such signs are difficult to produce, but are they difficultto read? Again, there exists littleevidence to indicate otherwise, since a heavily pictorial quality may actually facilitate the direct comprehension of meaning (john Monaghan, personal communication, 1992). Who Could Read and Write? A final topic concerns the number of people who could read or write in the Classical and pre-Modern world. For all societies the exact numbers are subject to dispute, yet most authors agree on one point: until relatively recent times, reading and writing accorded with a pattern that Havelock described as "craft literacy" (Havelock 1982: 1O, 187-88; Harris 1989: 7-8; see also Parsons 1966: 51). Craft literacy means that reading and writing were the preserve of a relatively small group that excluded most farmers, women, and unskilled laborers; as a skill, it was concentrated in the hands of the elite and of the craftsmen under their control. William Harris would argue that even during the peaks of Classicalliteracy no more than 1Oto 15 percent of the population could read and write (Harris 1989: 328). This situation only changes when strong forces, such as the introduction of the printing press, the reduction of dispersed rural settlement, and the introduction of philanthropic or state support, would begin to encourage literacy, particularly at the level of primary education (Harris 1989: 15). Nonetheless, the assumptions that underlie such conclusions are in themselves problematic. In this essay we have examined several themes: (1) the wide range of abilities in the production of, and response to, writing; (2) the close interaction between oral and written communication in most pre-Modern societies, regardless of the script involved; and (3) the understanding that most writing systems are less inhibiting and far more flexible and efficient than the partisans of alphabets would recognize. !lI!!! Stephen Houston 34 Toput these themes another way, and in answer to the questions posed above, literacy should be defined within scribal traditions and in variable ways; writing often fails to convey a complete message, but rather standsin a collaborative relationship with acoustic and bodily messages (seeMonaghan, in this volume); and technological determinism is not a usefulconcept in studies of ancient writing, which must be understood above all in its social dimensions. These observations do not diminish the value of careful work by scholars like Harris, who shows admirable restraintin his quantification of what is essentially qualitative evidence. Butthey do suggest that the low, if approximate, numbers proposed for the number of literates have several conceptual defects: they blur useful distinctions between reading and writing on the one hand and, on the other, fail to distinguish levels of competence in scribal production and response. The result may be a serious underrepresentation of the extent and impact of literacy.12 Maya literacy To date, there have been few discussions of Maya literacy. The subject is a difficult one, with even less pertinent information than exists for Old World scripts. Nonetheless, we do have four possible sources of evidence: the comparative information discussed thus far, which gives us a range of possibilities and perspectives; comments by Colonial observers about writing in Mesoamerica and in the Maya region; historicallinguistics; and hints from the Pre-Columbian period, particularly the Classic period (ca. A.D. 250 to 900), when, to judge from the number of texts, literacy was at its height. Colonial Evidence Among the most trenchant comments are those by Eric Thompson. He saw Maya writing as an aide memoire for learned discussions by priests about astronomical matters (Thompson 1971: 19,20,27); Maya glyphs were not for the masses, but were rather restricted to members of the nobility (Thompson 1972: 13; Villagutierre Soto-Mayor 1983: 224) and supervised closely by priests at the few centers of learning where books were produced (Tozzer 1941: 27).13This remark parallels statements by colonial writers (Tozzer 1941: 28; Thompson 1972: 13), who mention the limited extent of writing, which seldom entered the world of litigation and practical affairs (Tozzer 1941: 231 ).14On the authority of Gaspar Antonio Chi, Diego de Landa, Sanchez de Aguilar, and Bartolome de Las Literacy among the Maya 35 refers direct! Casas, Thompson also notes the close relationship between reading and writing to oral recitation, observing that historical tales and other nar- in a numbe: the sculptuiI . . IIteracy: If ' it signified 1989:196 of writing can easily has more, images th; site of Pie ratives were often sung or paralleled by oral disquisition-that is, they conformed to a pattern of recitation literacy (e.g., Tozzer 1941: 153).'5 Similar practices existed in Mexico in the years just before the Conquest: "Interpretation of the painted codices depended on intensive memorization by select groups. . . [and] . . . transmission of the extensive bodies of hymns, poems, and chronicles was essentially oral," although the written word represented an especially prestigious channel of information (Kartunnen 1982: 399; see also Sahagun 1950-1982,3:67; 1989: xxvi; Monaghan 1990). Gillespie anyone cj Anoth itself. G hanced t Thompson makes another comment as well: ". . . I feel that [the Colonial writers] refer to a full mastery of the subject. . . but there was surely another group with a limited knowledge of the writing. They might be termed semi-literate" (Thompson 1972: 13). That is, there existed a range say, the' also ma mindful eracy.' abouts thus in The chang' Initial! of literacy, including some people who, rather like those at Fayum, could barely write and read (Thompson 1971: 27). Historical Linguistics Cecil Brown uses historical linguistics to draw his conclusions hieroglyphic about literacy (1991: 495). He suggests that, because the Maya term for "writing" is more broadly diffused than the heterogeneous set of terms for "reading," most people could recognize writing, but could not read. Although ingenious, Brown's evidence does not justify his conclusions about ancient literacy. If people saw writing, and thus devised a term for it, they could also have devised a word for reading, which presumably they also witnessed, at least in the public contexts described by Bishop Diego de Landa (Tozzer 1941: 153). The diversity of terms seems less to represent a condition of rampant Pre-Columbian suc vers illiteracy than the extended process of reading in a society with recitation literacy: first, the reader "saw" the paper, then "counted" or construed the signs, and finally "spoke" or "called aloud" from the written page.'6 Thus, wa~ perij ing nUl Brown's reasoning fails to establish the extent of Pre-Columbian reading and writing, although it does support the presence of recitation Jiteracy among the ancient Maya. Evidence from the Pre-Columbian Period Direct evidence from the Pre-Columbian, period is not only the most speculative and especially the Classic but the most relevant, for it Stephen Houston 36 Md is ~ I w! di; refers directlyto the question at hand. The evidence has been dealt with in a numberof ways. Dorie Reents-Budet, for example, contends that thesculpturalpresentation of writing may indicate relative degrees of literacy:if the text of a monument faced away from the public, then it signifieda small audience for literate communication (Reents-Budet 1989:196). Her emphasis on the architectural and sculptural placement of writingis a useful one, although it should be stressedthat a viewer caneasilywalk around a monument, and that the positioning of texts hasmoreto do with the order of presentation, ranking, and decorum of imagesthanwith literacy (Baines 1983). One stela at the Classic Maya siteof PiedrasNegrashas hieroglyphs on its top, where it is doubtful that anyonecould read them (at least not without a ladder!). Anotherapproach is to examine the characteristics of Maya writing itself.GeorgeKubler believes the pictorial quality of Maya glyphs enhancedtheirreadability, sothat even "farmers in the field" could discern, say,the namesof rulers (Kubler 1973: 162; ]usteson 1986: 453). I have alsomadethis argument about public monuments (Houston 1989: 25), mindful,like Thompson, that there may have been many levels of literacy.!7My conclusion is precisely the opposite of Havelock's remarks aboutsyllabicscript, which Havelock saw as unnecessarily complex and thusinimical to fuliliteracy.!8 Thedevelopment of Maya script may also afford some clues about changingdegreesof literacy within the trajectory of the scribal tradition. Initially,Maya writing consisted of a linear series of logographs, each occupyinga glyph block. Within a few years scribes began to provide phoneticclues, or syllabic reinforcement for the logographs. Through thisdevice,they not only diminished the ambiguity of the logographs by signalingmore precisely their reading, but also made an indirect statementaboutliteracy: presumably, a larger circle of readerswould require suchcluesto an extent unnecessary among a smaller group of literates versedin the obscurities of the writing system. Or, to put it another way,the increaseduse of phonetic clues by the beginning of the Classic period,and, increasingly, purely phonetic spellings reflected not a writingsystemof heightened inaccessibility, but the existence of a growing numberof literates during the course of the Classic period. Parenthetically,a strong case can be made that the development of Mayascript responded lessto the unilineal sequencesfavored by Ignace Gelbthanto changing social and cultural conditions. For example, there is no clear evidence of an early stage of pictographic or ideographic writing in Mesoamerica (see Monaghan, in this volume). From all indications,the initial stages were exclusively logographic; virtually all Literacy among the Maya 37 ro-- to b' syllabic signs originated in logographs with weak consonants (justeson 1989: 33). I would even suggest that the whole concept of ideographs mul~ co needs revision. The few such signs that have been identified in Maya script may be characterized instead as "extended logographs," insofar B~ pOS! I as they convey words or phrases that are longer than the ConsonantVowel-Consonant-Vowel-Consonant (CVCVC) form contained in most spol me1 logographic signs. These signs are both rare and late because of the in- ca herent difficulties in reading them. As for purely ideographic writing, trQ I agree with A. Leroi-Gourhan that "true pictography is recent, mostly dating from after the period of contact with literate societies" 1987: 301; DeFrancis 1984: 133-148). There are a few archaeological 1~ th (Goody clues about the extent of Maya reading a; 0 and writing during the Pre-Columbian period. Perhaps the most direct evidence stems from the number of texts, since presumably a highly literate society would leave more written material than one with little emphasis on writing. Most epigraphers would accept that there survive, as a very rough estimate, between 7,500 and 10,000 hieroglyphic texts, including those on ceramic vessels. Of these, the majority date to the Late Classic period (ca. A.D. 600 to 850), when, at its peak, approximately three million people lived in the southern part of the Yucatan peninsula, where most of the texts occur (Turner 1990: 309). Because of sampling problems and the inherent difficulties of arguing from negative evidence, it is unlikely that the ratio of one text per three hundred people represents a significant figure for understanding Maya literacy (see Boring 1979: 6; Johnston 1983: 66). John Justeson has discussed a related approach: the use of mortuary offerings, some embellished with hieroglyphs, to "test whether elite persons were systematically more literate than commoners" (justeson 1978: 320-325). Yet the argument contains a crucial flaw, which Justeson him- self recognizes (justeson 1987: 324,326). The presence of writing in a grave cannot be regarded as solid evidence for the literacy of the deceased; as a prestige item, examples of script would be logical offerings in such contexts, regardless of the abilities of the perso(l interred in the grave. Nonetheless, the nature of such texts, which have been shown to be dedicatory or proprietary (Houston, Stuart, and Taube 1989), display an intriguing parallel with the earliest alphabetic writing in Greece and the central Mediterranean. In much of the Old World, the use of formulaic expressions may reveal both the cadences of recitation literacy and the limitations of the readers (Robb 1978: 32; Boring 1979: 1; Johnston 1983: 67; Havelock 1986: 85; Wallace 1989: 123-125). To explain: the formulae contain elements that are readily recognizable, although, Stephen Houston 38 to besure,with some virtuosic flourishes; the very existence mulaemayimply the need to accommodate of the for- those with marginal reading comprehension. Butperhapsthe best way of evaluating the extent of writing, as op- posedto reading, is by examining graffiti. These are marks that result from spontaneous, unforced acts, often done furtively; as such, they not only measuremore directly the ability to write but register deviations from canonicalspelling (see Wright 1976). In the Classical troveof graffiti has been used to examine world, the extent of writing the rich (Cartledge 1978:32)-the greater the number of dedicatory and onomastic graffiti, thegreaterthe number of writers (Stoddart and Whitley Inthe Mayaarea there occur many graffiti, although, a smallnumberhave been recorded, of Tikal (Graham 1967; Kampen 1988: 763). regrettably, only and most of these come from the site 1978; Trik and Kampen 1983).'9 The numbersaresuggestive.At Tikal there appear, as an extremely conservativeestimate,some 512 distinct graffiti; 26 of these, percent,are glyphic, with most displaying or approximately an extreme degree andbrevity(Trik and Kampen 1983), and some showing 5 of crudity ersatz spellings (Orrego andLarios1983)20 At the site of Rio Bec, which lies in an area strikingfor its paucity of monumental texts, there occur 95 graffiti, whichnoneare glyphic (Stoll 1979). From this we may deduce differencesin writing ability that are analogous betweenthe number of Attic and Cretan to the extreme graffiti (Stoddart 1988: 763). Yet even at Tikal the few graffiti appear of regional divide and Whitley to indicate the limitedextentof writing among the Classic Maya. The uncertainty that envelopsthis issue is unlikely to be overcome evidence:most graffiti at Tikal seem to be quite by the limitations of the late-circa AD. 850- 900(Kampen1978: 169), and thus may not reveal much of value about theextentof writing at the height of the Late Classic period. Conclusions Thispaperhas reviewed treatments of literacy in the ancient world and usedthesestudiesto understand Pre-Columbian Maya literacy. Its con- clusions:that reading and writing should be treated separately, that literacyshiftedwithin a person's lifetime and within the trajectory of a scribal tradition,and that scripts were often ancillary to recitation and performance. Moreover, a focus on the social context of literacy replaces the emphasisthat Havelock and others applied to the intrinsic consequences of certainwriting systems, although, to be sure, script does record malion"far beyond the carrying capacity of human memory, infor- individual Literacy among the Maya 39 \ 2 or collective" (Halverson 1992: 315) and as such represents a signal advance in human communication. A pretat high s contr As for the Maya evidence, all scholars agree that literacy must have been limited at all periods. But disagreement continues about the more subtle questions. I believe Maya writingdeveloped in ways that reflected increased literacy, particularly from the EarlyClassic on, although from various clues it seems that writing and perhaps reading were still restricted to relatively few people. Writing apparently varied in space, some areas having a vibrant tradition of writing and others, especially small communities but also whole regions, having none. The implications of this are unclear. Perhaps parts of the Maya region did not need glyphs to the same extent (see above for discussion about Sparta), a pattern suggesting great regional differences in the social and political contexts of literacy. Another explanation for such variability has less to do with what people needed-with what they did or did not want-than with the fact that others may have made such decisions for them. There exist a few scraps of evidence that monumental writing was under close political control; at the pleasure of the overlord, texts were meted out in small doses to subordinates at subsidiary centers (Houston 1993a). The site of Piedras Negras may even contain direct evidence of the centralized schools that supported this system, which lay embedded within a framework of political control {Satterthwaite1965). In regard to reading, I agree with Thompson and Kublerthat the pictorial features of the writing were maintained-that is, prevented from achieving the abstraction that characterizes Chinese logographs-by the need to preserve superficial reading ability among a larger group of people. However, any reasonable person would acknowledge how tenuous these assumptions are. Further work on graffiti, still much neglected, will provide some hope of elucidating Pre-Columbian literacy, as will a renewed search for the roles of writing in the ancient world. tiquit biogr ~~: \! I Wrltt J state I McKi I ]I , 3 span positJ; 4 acy i the e! tige; that: 332}, andl the i cou. CIa] COI Th pir: ab! 1. a Notes In preparing this paper I have benefited from the comments of Michael Coe, Mary Miller, John Monaghan, and the participants David Stuart, Norman Yoffee, an anonymous in the Dumbarton Oaks roundtable reviewer, on writing, organized by Elizabeth Boone in 1991. I thank Elizabeth for asking me to attend and for providing the opportunity to publish my thoughts. 1 Logosyllabic scripts are those that combine "word signs" (often a single mor- pheme) and "syllabic signs" (often a consonant and a vowel). In Maya script, the syllabic or "phonetic" signs derive from logographs (Houston and Stuart 1992a). Stephen Houston 40 \ r 2 An example from Medieval Europe illustrates one such problem of interpretation.According to Franz Bauml (1980), many people, particularly those of highsecular rank, were unlikely to have been literate, despite statements to the contrary.Rather, authors and biographers simply took their cue from Late Antiquity and particularly from the formulaic descriptions presented in the imperial biographiesof Suetonius (BaumI1980: 240). Such problems aside, Rosamond McKitterick (1989: 272) posits the existence of widespread literacy during the Carolingian period. In her opinion, Christianity established the centrality of the written word by emphasizing the special quality of authoritative texts. Although McKitterick is likely to be correct in her general estimate, Biiuml shows that statementsabout literacy are not always what they seem. 3 In both cases the signatures are qualitative measures of production and response.Eventually, quantitative values might be assigned to particular coordinate positionson the chart, although there is insufficient space to do so here. 4 In his reflections on Medieval reading and writing, Franz Bauml defines literacyin terms of social and political advantage (1980: 243). Those with access to the essentialknowledge transmitted by writing would gain distinction and prestige;those who did not enjoy such access would suffer disadvantage. The view thatwriting facilitates political and social hegemony is an old one (Harris 1989: 332).Other scholars would go further and place literacy within a system of control and exploitation, usually exercised by a centralized authority that apportioned the privilege of reading and writing with the aim of monopolizing political discourse(Levi-Strauss1955: 342-344; Wheatley 1971: 377; Morison 1972: 2, 14; Clanchy 1979: 263; Larsen 1988): "il faut admettre que la fonction primaire de la communication ecrite est de faciliter I'asservissement" (Levi-Strauss 1955: 344). Theuseof scripts to facilitate elite control may explain why, as in the Roman empire or in Tokugawa, Japan, the illiteracy of the masseswas thought to be a desirablecondition, since it "contributed to the stability of the political order" (Dore 1965: 215; Harris 1989: 333). Nonetheless, an emphasis on the political dimensions of writing can be taken too far. From ancient Egyptian evidence, John Baines deduces a complex, interactive relationship between writing and political control (Baines 1988: 208). With the exception of public monuments, the state could not always supervise the limits or consequences of literacy, which could veer off in ways difficult to contain. In general, the "statist" view of writing-the perspective that script serves solely as a means of political legitimation or control-fails to recognize the mul- tiple functions of writing in both the Old and New Worlds (pace Santley 1989: 96-97). 5 Although Havelock does not mention it, a similar tension between oral com- position and written redaction characterizes the epic literature of early Ireland Uackson1971 : 28). 6 The best-known example of this is Augustine's astonishment in watching Ambrose at his books, reading silently. Yet this illustration of ancient reading Literacy among the Maya 41 , I should be taken with a grain of salt: after all, Augustine was a poor African pro- as vincial, Ambrose the child of sophisticated and wealthy parents from the center of the empire (Knox 1968: 422). Perhaps silent reading was by then common, as Bernard Knox proposes for Athens during the fifth and fourth centuries B.C. (Knox 1968: 434). For a different view, see Lentz's discussion of silent reading in dramatic contexts (Lentz 1989: 163-164). 7 Brian Stock advances a similar hypothesis in regard to the advent of wide- :~~JI I spread, alphabetic literacy in northern Europe (Stock 1983: 18, 531). Not since 13 Classical Antiquity had human beings been able to distinguish philosophically between "outer versus inner, independent object as opposed to reflecting subject, sup! hi or abstract sets of rules in contrast to a coherent texture of facts and meanings" (Stock 1983: 531). The eventual result of these distinctions was the rebirth of ~8)1 hermeneutics in medieval clothing. 8 In discussing three successive works by Jack Goody, John Halverson charts the disintegration of Goody's variant of the "literacy thesis"-the idea that writ- ing instigated a profound cognitive rupture with an oral past (Halverson 1992). Halverson shows that logical syllogisms or abstract thought do not necessariIy stem from writing, and that, through time, Goody makes a progressively weaker case for such a connection. 9 "The native syllabary (in Japan)was something called onnade or the 'woman's hand' (otokomojii, 'men's letters', referring to Chinese characters)" (Morris 1964: 212). The references to female literacy in Heian Japanare noteworthy, for there is a consensus from other parts of the pre-Modern world that relatively few women could read or write (Harvey 1966: 621; Rawski 1979: 22-23; Harris 1989: 329); those that could would have belonged to only "the very best families" (Harris 1989: 280), although direct support for this assumption is elusive. I should mention that one reader of this paper objects to "limited female literacy" as an idea smacking of androcentric bias and weak attention to alternative possibilities. Yet the Maya case is instructive: in Classic art and text women are not once associated with the accoutrements or titles of scribes. We should remember that the issue here is not so much innate ability, as the social conditions that enable literacy. 10 Nonetheless, there are other views about the ease with which people can achievefunctionalliteracyin Chinesescript.DeFrancis(1984:205) strongly questions Rawski's inflation of the numbers of people who could read or write. Rather, like Havelock, he contends that Chinese script is inherently difficult to learn and to retain, making it unsuitable for the needs of modern, literate society (DeFrancis 1984: 220; 1989: 243). Yet we should also remember that DeF.rancis spent many years teaching Chinese to American students (DeFrancis1984).Onewonders whether this experience colored his perception of the intrinsic difficulties of Chinese writing. 11 Part of the problem stems from the conception of writing as a technology, as something comparable to the steam engine in its dramatic consequences for society (Havelock 1982: 6; Ong 1982: 81-83). True, writing can be described L:I'=o'~"'- Stephen Houston 42 I as a manual skill with specific objectives-i.e., servesas a means of communication The use of the technological metaphor that allows a technology~but production is regrettable, above all it and elicits response. since it separates script from other kinds of communication. 12 Since our emphasis has been on the cultural it might be best to adopt the general sense of "craft and social context literacy" label, which remains firmly rooted in the artistic production of literacy, but not the specific of Classical Antiquity. 13 Bishop Landa specifically mentions that books were produced under the supervisionof the highest-ranking priest and then distributed with minor priests whentheytraveled to smaller communities to discharge their duties (Toner 1941 : 28).This suggests,first, that schools were not common and, second, that there wererelatively few places where writing was produced. 14 However, this does not seem to have been the case in Post-Classic Central Mexico, where writing had wider uses. According to Fray Bernardino de Sahagun,writing was employed by horticulturists, soothsayers, physicians, and evenjudges(Sahagun 1950-1982, 8: 55; 10: 29, 31,42). The contrast with Colonial descriptions of Maya script points to one of two possibilities: that Colonial authorsinaccurately emphasized the limited usesof Maya script; or that Colonial writers fairly portrayed the functional differences between Mesoamerican writing systems.Accordingly, the Mexica Aztec-governed by a bureaucratic state- mighthavehad broader uses for writing than did the Post-ClassicMaya, who lived within segmentary states of negligible size and bureaucratic development. 15 In precontact Yucatan there existed a close connection between song, dance, and political gatherings. Important officials known as the Hol Pop both presided over deliberative bodies (Barrera Vasquez 1980: 228) and supervised dances, singing,and the storage and playing of musical instruments (Toner 1941: 93). 16 David Stuart points out that "seeing" may indeed have been the expression for reading during the Classic period, when rulers referred to the act of "seeing" carving stones (personal communication, 1992). Other criticisms of Brown's paper appear in Tedlock (1992), with further comments on Brown and Tedlock by Houston and Stuart (1992b). 17 Cecil Brown construes my remarks to mean that I propose widespread liter- acyduring the Classic period (Brown 1991: 25). My intention was rather to show that the pictorial nature of Maya glyphs had an effect on reading, since it gave to a wider group of people the superficial ability to understand parts of the script. But I also believe that the entire text, and especially its more recondite features, could be produced by only a few people. The crucial distinction, therefore, is between reading and writing. In my opinion, these are categories that must remain separate. 18 Eric Thompson saw a close relationship between the mindset of the ancient Maya and their script. Unlike Havelock, Thompson believed that writing reflected ratherthan triggered features of thought (Thompson 1971: 15). He did, however, Literacy among the Maya 43 I l1li I"""'""' feel that Maya script was sui generis, a unique development that did not com- Civil, RevU4 pare easily with other writing systems (Thompson 1971: 15). Perhaps this is why Thompson responded so negatively to Yurii Knorosov's studies, which drew upon structural parallels between Maya glyphs and syllabic scripts of the Old World (Knorosov 1958). 19 Karl Taube informs me that graffiti are relatively common in parts of northern The, Yucatan, although few of these marks have been recorded (personal communication, 1992). There is an urgent need for graffiti to be documented, since many Neb1 DeF versi surfaces with Pre-Columbian markings have been heavily damaged by tourists. At Tikal, such wanton destruction mars most of the surviving graffiti. It is also surprising how few accounts there are of graffiti in other parts of Mesoamericacould it be that people have simply failed to notice them? lulu 20 Do Another axis of variability is the fact that some buildings have a great deal of graffiti, others little or none. A fuller survey of graffiti in the Maya Lowlands might establish correlations between abundance of graffiti and building type. References Baines, John. 1983. Literacy and Ancient Egyptian Society. Man 18: 572~599. -. 1988. Literacy, Social Organization, and the Archaeological Record: The Case for Early Egypt. In State and Society: The Emergence and Development of Social Hierarchy and Centralization, ed. John Gledhill, Barbara Bender, and Mogens Trolle Larsen, 192-214. London: Unwin Hyman. Balogh, E. 1927. Voces paginarum: Beitrage zur Geschichte des lauten Lesens und Schreibens. Phi/ologus 82: 84-109, 202-240. Barrera Vasquez, Alfredo, ed. 1980. DiccionarioMaya Cordemex, Maya-Espanol, Espanol-Maya. Merida: Ediciones Cordemex. Bauml, Franz H. 1980. Varieties and Consequences of Medieval Literacy and Illiteracy. Speculum 55: 237-265. Boring, Terrence A. 1979. Literacyin Ancient Sparta. Leiden: E.J. Brill. Brown, Cecil L. 1991. Hieroglyphic Literacy in Ancient Mayaland: Inferences from Linguistic Data. Current Anthropology 32: 489-496. Cartledge, P. A. 1978. Literacy in the Spartan Oligarchy. journal of Hellenic Studies 98: 25-37. Chafe, Wallace, and Deborah Tannen. 1987. The Relation between the Written and the Spoken Word. Annual Review of Anthropology 16: 383-407. Cipolla, Carlo M. 1969. Literacyand Development in the West. Harmondsworth: Penguin. Civil, Miguel. 1972. The Sumerian Writing System: Some Problems. Orientalia 42: 21-34. ~ Stephen 44 Houston ,.. Civil, Miguel, and R. D. Biggs. 1966. Notes sur des textes sumeriens Revue d'Assyriologie archa'iques. 60: 1-16. Clanchy, M. T. 1979. From Memory Cambridge, Massachusetts: Harvard to Written University Record: England 1066-1307. Press. Cross,Frank Moore. 1989. The Invention and Development of the Alphabet. In The Origins of Writing, ed. Wayne M. Senner, 77-90. Lincoln: University of Nebraska Press. DeFrancis, John. 1984. The Chinese Language: Fact and Fantasy. Honolulu: Uni- versity of Hawai i Press. -.1989. Visible Speech: The Diverse Oneness of Writing Systems. Hono- lulu: University of Hawaii Press. Doane,A. N. 1991. Introduction. In Vox intexta: Orality and Textuality in the Middle Ages, ed. A. N. Doane and Carol Braun Pasternack, xi-xiv. Madison: Universityof Wisconsin Press. Dare, Ronald P. 1965. Education in Tokugawa Japan. London: Routledge and Kegan Pau I. Finnegan,Ruth. 1988. Literacy and Orality: Studies in the Technology of Communication. Oxford: Basil Blackwell. Gelb, Ignace J. 1963. A Study of Writing. 2d ed. Chicago: University of Chicago Press. Gillespie, Susan D. 1989. The Aztec Kings: The Construction of Rulership in Mexico City. Tucson, Arizona: University of Arizona Press. Goody, Jack. 1977. The Domestication bridge University -. of the Savage Mind. 1987. The Interface Cambridge University Cam- Between the Written and the Oral. Cambridge: Press. Goody, Jack, and Ian Watt. 1963. The Consequences Studies in Society and History Gough, Kathleen. 1968. Implications of Literacy. Comparative 5 (3): 304-345. of Literacy in Traditional Literacy in Traditional Societies, ed. Jack Goody, 69-84. University Cambridge: Press. China and India. In Cambridge: Cambridge Press. Graham, Ian. 1967. Archaeological American Research Institute, Halverson, John. 1992. Goody Explorations in EI Peten, Guatemala. Middle Publication 33. New Orleans: and the Implosion Tulane University. of the Literacy Thesis. Man 27: 301-317. Harris, William V. 1989. Ancient Literacy. Cambridge, Massachusetts: Harvard University Press. Harvey, F. D. 1966. Literacy in the Athenian Democracy. Revue des Etudes Grecques 79: 585-635. Literacy among the Maya 45 r : Havelock,EricA. 1982. The Literate Revolution in Greece and Its Cultural Consequences. Princeton: Princeton University Press. -. 1986. The Muse Learns to Write: Reflections on Orality and Literacy from Antiquity to the Present. New Haven: Yale Uni:~sity Press. Heath, S. B. 1986. Literacy and Language Change. Languages and Linguistics: The Interdependence of Theory, Data, and Application, Georgetown University Round Table on Languages and Linguistics, 7985, ed. D. Tannen and j. E. Alatis, 282-293. Washington, D.C.: Georgetown University Press. Houston, Stephen D. 1989. Reading the Past: Maya Glyphs. Berkeley: University of California Press. -. Lar! 1993a. Hieroglyphs and History at Dos Pilas: Dynastic Politics of the Classic Maya. Austin: University of Texas Press. Stal -. Pol: 1993b. Mesoamerican Writing. In The Encyclopedia of Language and Linguistics. Oxford: Pergamon Press, in press. Lar' Houston, Stephen D., and David Stuart. 1992a. Logogramas, fonetismo y la natu- Lei raleza de la escritura Maya. Historia general de Guatemala Tomo 1. Guatemala: So Fundacion para la cultura y desarollo, in press. Le Lo 1992b. On Classic Maya Literacy. Current Anthropology 33 (5): 589593. on Houston, Stephen D., David Stuart, and Karl Taube. 1989. Folk Classification of Mi Classic Maya Pottery. American Anthropologist 91: 720-726. 2d jackson, Kenneth H. 1971. A Celtic Miscellany: Translations from Celtic Litera- d tures. Harmondsworth: Penguin. johnston, A. 1983. The Extent and Use of Literacy: The Archaeological Evidence. M In The Greek Renaissance of the Eighth Century BC: Tradition and Innovation, AI ed. Robin Hagg, 63-68. Stockholm: Skrifter Utgivna av Svenska Institutet i Athen. Justeson, ). 1978. Mayan Scribal Practice in the Classic Period: A Test-case of an Explanatory Approach to the Study of Writing Systems. Ph.D. diss., Department of Anthropology, Stanford University. Ann Arbor: University Microfilms. -. MI 1986. The Origin of Writing Systems: Preclassic Mesoamerica. World ]9 Archaeology 17: 437-458. -.1989. L The Representational Conventions of Mayan Hieroglyphic Writ- ing. In Word and Image in Maya Culture: Explorations in Language, Writing, and Representation, ed. William F. Hanks and Don S. Rice, 25-38. Salt Lake City: University of Utah Press. Kampen, Michael. 1978. The Graffiti of Tikal. Estudios de Cultura Maya 5: 155180. Kartunnen, Frances. 1982. Nahuatl Literacy. In The Inca and Aztec States, 74007800, ed. George A. Collier, Renato I. Rosaldo, and j. D. Wirth, 395-417. New York: Academic Press. -- Stephen Houston 46 ,.. Keightley, DavidN. 1989. The Origins of Writing in China: Scripts and Cultural Contexts.In The Origins of Writing, ed. Wayne M. Senner, 171-202. Lincoln: Universityof Nebraska Press.' Knorosov,Yurii V. 1958. The Problem of the Study of the Ma;~.HieroglyphiC System.American Antiquity 23: 284-291. Knox, Bernard M. W. 1968. Silent Reading in Antiquity. Greek, Roman, and ByzantineStudies9: 421-435. Kubler,George. 1973. The Clauses of Classic Maya Inscriptions. In Mesoamerican Writing Systems,ed. Elizabeth P. Benson, 145-164. Washington, D.C.: DumbartonOaks. Larsen, MogensTrolle. 1988. Introduction: Literacy and Social Complexity. In and Society: The Emergence and Development of Social Hierarchy and PoliticalCentralization, ed. John Gledhill, Barbara Bender, and Mogens Trolle State Larsen,173-191. London: Unwin Hyman. Lentz, Tony M. 1989. Orality and Literacy in Hellenic Greece. Carbondale: SouthernIllinois University Press. Levi-Strauss,Claude. 1955. TristesTropiques. Paris: Librairie Pion. Logan,RobertK. 1986. The Alphabet Effect: The Impact of the Phonetic Alphabet on the Development of Western Civilization. New York: William Morrow. Mann,). C. 1985. Epigraphic Consciousness.journal of Roman Studies 75: 204206. McKitterick, Rosamond. 1989. The Carolingiansand the Written Word. Cambridge:Cambridge University Press. Monaghan,John. 1990. Performance and the Structure of the Mixtec Codices. Ancient Mesoamerica 1 : 133-1 40. Morison, Stanley. 1972. Politics and Script: Aspects of Authority and Freedom in the Development of Graeco-Latin Script from the Sixth Century 8.C. to the Twentieth Century A.D. Ed. N. Barker. Oxford: Clarendon Press. Morris, Ivan. 1964. The World of the Shining Prince:Court Life in Ancient japan. Oxford: Oxford University Press. Ong, Walter J. 1982. Orality and Literacy: The Technologizing of the Word. London: Methuen. Orrego Corzo, Miguel, and Rudy Larios Vallalta. 1983. Reporte de las investigaciones en el Grupo 5E-7 7, Tikal, Peten. Guatemala: Parque Nacional Tikal, Instituto de Antropologia e Historia de Guatemala. Parry, Milman. 1971. The Making of Homeric Verse: The Collected Papers of MilmanParry.Ed. Adam Parry. Oxford: Clarendon Press. Parsons,Talcott. 1966. Societies: Evolutionary and Comparative Perspectives. New York: Prentice-Hall. II Literacy among the Maya 47 ..- . Ti Rawski, Evelyn Sakakida. 1979. Education and Popular Literacy in Ch'ing China. Ann Arbor: University of Michigan Press. Reents-Budet, Dorie. 1989. Narrative in Classic Maya Art. In Word and Image in Maya Culture: Explorations in Language, Writing, and Representation, William F. Hanks and Don S. Rice, 189-197. Utah Press. ed. Salt Lake City: University of Robb, Kevin. 1986. Poetic Sources of the Greek Alphabet: Rhythm and Abecedarium from Phoenician to Greek. In Communication Arts in the Ancient World, ed. Eric A. Havelock and Jackson P. Hershbell, 23-36. New York: Hastings House. Sahagun, Fray Bernardino de. 1950-1982. Florentine Codex: General History of the Things of New Spain. Trans. Arthur J. O. Anderson and Charles E. Dibble. Salt Lake City and Santa Fe: University of Utah Press and School of American Research. Sampson, Geoffrey. 1985. Writing Systems: A Linguistic Introduction. Stanford: 0: Stanford University Press. Sanderson, Michael. 1972. Literacy and Social Mobility in the Industrial Revolution. Past and Present 56: 75-104. Santley, Robert 5.1989. Writing Systems, Political Power, and the Internal Struc- ture of Early States in Precolumbian rent Archaeological Chacmool, Proceedings Archaeological Satterthwaite, Mesoamerica. In Cultures in Conflict: Cur- Perspectives, ed. Diana Claire Tkaczuk and Brian C. Vivian. of the Twentieth Annual Conferences, 90-99. Calgary: Association of the University of Calgary. Linton. 1965. Maya Practice Stone-Carving at Piedras Negras. Ex- pedition (Winter): 9-18. Schofield, R. S. 1968. The Measurement of Literacy in Pre-Industrial England. In Literacy in Traditional Societies, ed. Jack Goody, 311-325. Cambridge: Cam- bridge University Press. Scallon, R., and S. B. K. Scallon. 1981. Narrative, Literacy, and Face in Interethnic Communication. Norwood, New Jersey: Ablex. Stock, Brian. 1983. The Implications of Literacy: Written Language and Models of Interpretation in the Eleventh and Twelfth Centuries. Princeton: Princeton Uni- R versity Press. Stoddart, Simon, and James Whitley. 1988. The Social Context of Literacy in Archaic Greece and Etruria. Antiquity 62: 761-772. Stoll, Russell F. 1979. Lowland Maya Graffiti: Including a Presentation of the Graffiti at Rio Bec Temple B. Master's Thesis, Division of Graduate Studies, University of the Americas, Cholula, Puebla, Mexica. Street, Brian V. 1984. Literacy in Theory and Practice. Cambridge: Cambridge University Press. Tedlock, Dennis. 1992. On Hieroglyphic Literacy in Ancient Mayaland: An Alternative Interpretation. , Current Anthropology 33 (2): 216-218. Stephen Houston 48 ,. Thomas,Rosalind. 1989. Oral Tradition and Written Record in Classical Athens. Cambridge:Cambridge Unviersity Press. Thompson,j. Eric S. 1971. Maya Hieroglyphic Writing: An Introduction. 3d ed. Norman:University of Oklahoma Press. -. 1972. A Commentary on the Dresden Codex: A Maya Hieroglyphic Book.Philadelphia: American Philosophical Society. Tozzer,Alfred Marston, ed. 1941. Landa's Relacion de las Cosas de Yucatan. Papersof the Peabody Museum of American Archaeology and Ethnology 18. Cambridge,Massachusetts:Harvard University Press. Trik,Helen, and Michael E. Kampen. 1983. The Graffiti ofTikal. In Tikal Report No. 31, ed. William R. Coe. University Museum Monograph 57. Philadelphia: TheUniversityMuseum, University of Pennsylvania. Troll, DeniseA. 1990. The Illiterate Mode of Written Communication: The Work of the Medieval Scribe. In Oral and Written Communication: Historical Ap- proaches,ed. Richard L. Enos, 96-125. Newbury Park: Sage Publications. Turner, B. L., II. 1990. Population 1000 B.C. to AD. Reconstruction 1500. In Precolumbian lands, ed. T. Patrick Culbert for the Central Maya Lowlands: Population and Don S. Rice, 301-324. History in the Maya Low- Albuquerque: University of New Mexico Press. VillagutierreSoto-Mayor, Juan de. 1983. History of the Conquest of the Province of the Itza. Culver City: Labyrinthos. Wallace, Rex. 1989. The Origins and Development of the Latin Alphabet. In TheOrigins of Writing, ed. Wayne M. Senner, 21-35. Lincoln: University of Nebraska Press. Wheatley,Paul. 1971. The Pivot of the Four Quarters: A Preliminary Enquiry Into the Origins and Character of the Ancient Chinese City. Edinburgh: Edinburgh University Press. Wright, R. 1976. Speaking, Reading and Writing Late Latin and Early Romance. Neophilologus 60: 178-189. Youtie, H. C. 1971. {3pOll)tm<; /,pac/>mv: Between Literacy and Illiteracy. Creek, Roman, and Byzantine Studies 12: 239-261. Literacy among the Maya 49 ~ CP = Collected xeroxes (# 1) S. Houston: 2004. ‘The Archaeology of Communication Technologies’ . Annual Review of Anthropology 33:223-250. (# 2) S. Houston: 2004. ‘Writing in Early Mesoamerica’. In The First Writing: Script Invention as History and Process, ed. Stephen Houston, 274-309. Cambridge: Cambridge University Press. (# 3) N. Grube: 1994. ‘Observations on the History of Maya Hieroglyphic Writing’. In Seventh Palenque Round Table, 1989,ed. by Virginia M. Fields, ed. pp. 177-186. San Francisco: Pre-Columbian Art Research Institute. (# 4) S. Houston, J. Baines, J. Cooper: 2003. ‘Last Writing: Script Obsolescence in Egypt, Mesopotamia, and Mesoamerica. Comparative Studies in Society and History 45(3):430-480. (# 5) S. Houston: 1994. ‘Literacy Among the Precolumbian Maya: A Comparative Perspective’. Writing Without Words: Alternative Literacies in Mesoamerica and the Andes, edited by E. Boone and W. Mignolo, pp.27-49. Durham: Duke University Press (#6) N. Grube: 2004. ‘The Orthographic Distinction Between Velar and Glottal Spirants in Maya Hieroglyphic Writing’. In The Linguistics of Maya Writing, ed by Søren Wichman, ed. pp. 61-82. Salt Lake City: University of Utah Press. (# 7) S. Houston, D. Stuart, J. Robertson: 2004. ‘Disharmony in Maya Hieroglyphic Writing: Linguistic Change and Continuity in Classic Society’. In The Linguistics of Maya Writing, ed. by Søren Wichmann, 83-101. Salt Lake City: U of Utah Press. (reprinted and correct version of 1998 article) (# 8) M. Zender: 2005. ‘The Raccoon Glyph in Classic Maya Writing’.The PARI Journal 5(4):6-16. (# 9) S. Houston, J. Robertson, D. Stuart: 2000. ‘The Language of Classic Maya Inscriptions’, Current Anthropology 41(3): 321-356. (# 10) S. Houston, J. Robertson, and D. Stuart: 2001. Quality and Quantity in Glyphic Nouns and Adjectives, Research Reports on Ancient Maya Writing, 47. Washington, D.C.: Center for Maya Research. (#11) A. Lacadena: 2004. ‘Passive Voice in Classic Mayan Texts: cv-h-c-aj and -n-aj Constructions’. In The Linguistics of Maya Writing. ed by. Soren Wichman, pp. 165-194. Salt Lake City: University of Utah Press. (#12) N. Grube: 1998. ‘Speaking Through Stones: A Quotative Particle in Maya Hieroglyphic Inscriptions. In 50 años de estudios americanistas en la Universidad de Bonn: Nuevas contribuciones a la arqueología, etnohistoria, etnolingüística y etnografía de las Américas., ed by Sabine Dedenbach-Salazar Sáenz, et al., pp. 543-558. Markt Schwaben: Verlag Anton Saurwein. (# 13) J. Justeson: 1989. ‘Ancient Maya Ethnoastronomy: An Overview of Hieroglyphic Sources’. In World Archaeoastronomy. ed. by Anthony F. Aveni, ed. pp. 76-129. Cambridge: Cambridge University Press. (# 14) N. Grube and S. Martin: 2000. ‘Dynastic History of the Maya’. In Maya: Divine Kings of the Rainforest, ed. by Nikolai Grube, pp. 148-171. Cologne: Könemann Verlagsgesellschaft. (# 15) S. Martin: 2003. ‘In Line of the Founder: A View of Dynastic Politics at Tikal’. In Tikal: Dynasties, Foreigners, and Affairs of State, ed. by Jeremy A. Sabloff, pp. 3-46. Santa Fe: School of American Research Press. (# 16) S. Martin: 2000. ‘At the Periphery: The Movement, Modification and Re-use of Early Monuments in the Environs of Tikal. In The Sacred and the Profane: Architecture and Identity in the Maya Lowlands, ed by P. R. Colas, K. Delvendahl, M. Kuhnert, and A. Schubart, pp. 51-61. Acta Mesoamericana, 10 Markt Schwaben: Verlag Anton Saurwein, (# 17) D. Stuart: 1998. 'The Fire Enters his House': Architecture and Ritual in Classic Maya Texts In Function and Meaning in Classic Maya Architecture, ed. by S. D. Houston, pp. 373-425. Washington, D.C.: Dumbarton Oaks. (# 18) D. Stuart: 1999. ‘Arrival of Strangers: Teotihuacan and Tollan in Classic Maya History’. Mesoamerica's Classic Heritage: Teotihuacán to the Aztecs. Davíd Carrasco, Lindsay Jones, and Scott Sessions, eds. pp. 465-513. Niwot: University Press of Colorado. (# 19) D. Stuart: 1997. Kinship Terms in Maya Inscriptions. In The Language of Maya Hieroglyphs, ed. by Martha J. Macri and Anabel Ford, pp. 1-11. San Francisco: Pre-Columbian Art Research Institute. (# 20) D. Stuart: “The Maya Hieroglyphs for Mam’, unpublished ms. (# 21) S. Houston and D. Stuart: 1996. ‘Of Gods, Glyphs, and Kings: Divinity and Rulership among the Classic Maya’. Antiquity 70:289-312.