Introducing the Eskaya Writing System: A Complex Messianic Script from the Southern Philippines

* This paper has benefited greatly from the input of other scholars, particularly the participants in the International Workshop on Endangered Scripts of Island Southeast Asia, hosted by the Linguistic Dynamics Science Project at the Tokyo University of Foreign Studies, 27 February–4 March 2014. Thanks especially to Asako Shiohara and Christopher Miller. I am indebted to Eskaya teachers for generously introducing me to the script, in particular Decena Nida Palma Salingay, Gaudencia Pizaña, Naning Galambao and Alberta Galambao. The Eskaya glyphs reproduced in this paper were developed with assistance from Marsiana Galambao (expert Eskaya consult), Siva Kalyan (consultant), and designers Mark Eastwood, Mikka Lagrimas, Joe Elvis de los Reyes, Michelle Gamboa, Lloyd Alden, Wap Martinez-Mercader and Cleobie Impang. Note that a font for the Eskaya script is only partially developed, although vector and image files are archived in PARADISEC.

Abstract

This paper introduces and documents the Eskaya () writing system of the Philippines, developed ca. 1920–1937, and attempts to reconstruct the circumstances of its creation. Although the script is used for representing Visayan (Cebuano)—a widely used language of the southern Philippines—its privileged role is in the written reproduction of a constructed utopian language, referred to as Eskayan or Bisayan Declarado. Held to have been invented by the ancestral ‘Pope Pinay’, the Eskayan language and its script are used by approximately 550 people for restricted purposes in the southeast of the island of Bohol. Of the approximately 1,065 characters in the system, a primary set of 24 are alphabetic with optional syllabic values; the remaining letters have syllabic values only and can be decomposed into an inahan (‘mother’), standing for (C)V, and a sinyas (‘gesture’) indicating consonant diacritics on either side of the nucleus. Coda diacritics are largely inconsistent, meaning that each syllabic character needs to be acquired independently. The script has minor logographic elements with ideography employed in the decimal numeral system. Over half of all Eskaya characters are redundant and at least 37 represent phonotactic impossibilities in either Visayan or Eskayan. The sheer size, complexity and irregularity of the hybrid Eskaya script is unparalleled among the world's writing systems. I argue that the very opacity of Eskaya writing is, in part, what makes it attractive to new learners and has contributed to its successful transmission for 90 years.

Keywords:

• Grammatogeny
• Writing Systems
• Philippine Scripts
• Artificial Languages

ORCID

Piers Kelly http://orcid.org/0000-0002-6467-2338

Notes

* This paper has benefited greatly from the input of other scholars, particularly the participants in the International Workshop on Endangered Scripts of Island Southeast Asia, hosted by the Linguistic Dynamics Science Project at the Tokyo University of Foreign Studies, 27 February–4 March 2014. Thanks especially to Asako Shiohara and Christopher Miller. I am indebted to Eskaya teachers for generously introducing me to the script, in particular Decena Nida Palma Salingay, Gaudencia Pizaña, Naning Galambao and Alberta Galambao. The Eskaya glyphs reproduced in this paper were developed with assistance from Marsiana Galambao (expert Eskaya consult), Siva Kalyan (consultant), and designers Mark Eastwood, Mikka Lagrimas, Joe Elvis de los Reyes, Michelle Gamboa, Lloyd Alden, Wap Martinez-Mercader and Cleobie Impang. Note that a font for the Eskaya script is only partially developed, although vector and image files are archived in PARADISEC.

¹Others include the enigmatic Raffles Script of Indonesia reported in 1817 (Raffles [1817] 1988), and the Leke script of Burma that was developed during 1844–1845 (Womack 2005). Note that Konrad Tuchsherer and P.E.H. Hair have investigated a potential historical connection between Cherokee and Vai (Tuchsherer & Hair 2002), but their findings are inconclusive.

²For an overview of new West African scripts see Dalby (1967, 1968, 1969), Schmitt (1980) and Unseth (2011); the early twentieth-century Asia-Pacific scripts that have come to my attention are the Caroline Islands script (Riesenberg & Kaneshiro 1960), the Khom script of Laos (Sidwell 2008), the Eskaya script described in this paper, the Mama script of Easter Island (Fischer 1997) and the Iban script of Malaysian Borneo (Philip 2007). In the latter half of the century, two independent scripts were developed for the Hmong language of Laos (Smalley et al. 1990; Smalley & Wimuttikosol 1998), and scripts are reported in Vanuatu (Gray 2012) and Bougainville (pers. comm. James Tanis, Ruth Spriggs, Steven Tamiung).

³This is the conventional estimate provided by Eskaya leaders and the National Commission on Indigenous Peoples in Bohol, although there are no survey or census data to support this claim. The only study to have provided a precise figure puts the population at 739 (Anania 2010) but it is not clear whether this number applies to all Eskaya people or only to those living within the municipality of Pilar.

⁴The Ethnologue estimate is drawn from census data that probably don't take into account second-language speakers of Visayan, particularly in Mindanao.

⁵The Eskaya Digital Archive hosted by the Pacific and Regional Archive for Digital Sources (PARADISEC) can be accessed at http://catalog.paradisec.org.au/collections/PK2/. Items of specific relevance to this paper include PK2-04-MANCAD02 (an Abidiha and Simplit belonging to Gaudenci Pizaña of Cadapdapan), PK2-04-MANCAD05 (a rare collection of Eskaya literature that includes a Romanized transliteration and a Visayan translation), PK2-03-MANBIAB09 (the expository text ‘Atikisis’) and PK2-03-MANBIAB12 (the Castañares Manuscript, one of the oldest surviving Eskaya documents).

⁶A version of ‘The Spanish and Eskayan Alphabets’ is archived at http://catalog.paradisec.org.au/collections/PK2/items/04, filename PK2-04-MANCAD05. The text begins on page 49. Note that the relationship between the Eskaya script and the human body has a parallel in one version of the Meetei Mayek script of North East India where consonant letters are ordered following a notional ordering of human body parts after which each letter is named (Singh 2011).

⁷Versions of the text ‘Atikisis’ are archived at http://catalog.paradisec.org.au/collections/PK2/, filenames PK2-03-MANBIAB09, PK2-03-MANBIAB10 and PK2-04-MANCAD01.

⁸I follow the precedent of Christopher Miller who uses the terms ‘Philippine script’ and ‘Old Philippine script’ to refer to the attested variants of the indigenous writing system of the Philippines (Miller 2011, 2014). The manner in which colonial chroniclers documented and named script samples by region, was, in the words of Juan R. Francisco, ‘an unconscious error, if not indeed a deliberate scheme, among earlier writers in their effort to create multiple cultural complexes in the Philippines. [ … ] upon examination of all these systems, there appears to be a singular affinity among them. If there was evidence of variety, this can only be understood as a result of the idiosyncrasies of the individual writers' (Francisco 1973).

⁹Within the conventions of this notation ‘G’ represents a glide, parentheses enclose optional values and / stands for ‘or’.

¹⁰This novena is led by the female members of the congregation every Sunday after Mass. It does not appear to follow the Roman Catholic novena tradition of a nine-day devotional observance. For more on Eskaya ideologies of writing see Kelly (forthcoming-a).

¹¹See Marciana Galambao's documentation of the Abidiha, numeral set and punctuation here: http://www.ling.hawaii.edu/ldtc/languages/eskaya/WritingSystem.html. Note that she has recorded monosyllabic Eskayan terms for punctuation marks and mathematical functions, potentially allowing meta-commentary in Eskayan on these aspects of the system.

¹²See http://catalog.paradisec.org.au/collections/PK2. The most comprehensive reference syllabary I have had access to is the item PK2-04-MANCAD02, penned by Gaudencia Pizaña. It has approximately 1,065 individual litri which are subdivided into 56 sets. (This figure excludes repetitions of identical characters in different sections of a Simplit but includes variant characters of the same syllable.) A smaller but beautifully illustrated version is PK2-05-MANTAY02 penned by Alberta Galambao of Taytay.

¹³A precedent for this system of vocalizing recited letter names is found in the Type 2 variant of Caroline Islands script (Riesenberg & Kaneshiro 1960).

¹⁴Here the written distinction between ‘ch’ and ‘chd’ is not an idiosyncrasy of the scribe. That the characters in Figure 6 feature both /ʤ/ and /tʃ/ sounds can be adduced through comparison with other reference syllabaries and in evidence from the corpus.

¹⁵Note nonetheless that the left-most graphic element of ‹chdiyaru› can at least be isolated as the syllable ‹chdi›. One Simplit includes a character for the sequence tsudub /tʃudub/ (see Fig. 6) which does not have a known meaning and is not attested in Eskaya literature or wordlists.

¹⁶See the document PK2-04-MANCAD.pdf pages 26–28, in Eskaya Manuscripts from Cadapdapan Bohol (http://catalog.paradisec.org.au/collections/PK2).

¹⁷Interestingly, Pahawh Hmong also has unique symbols for arithmetic functions, although the original Source Version did not include a zero (Smalley et al. 1990: 79).

¹⁸This is evident from etymologies of Visayan numerals but is also noted in the historical record. Ignacio Francisco Alcina observed that ‘[the Visayans] did not have arithmetic or numbers which may correspond to ours in writing, although, it is certain they counted by tens as we do' ([1668] 2005: 91).

¹⁹Jes Tirol argued that tri (‘two’), kuy (three’)and pan (‘four’) were derived from Sanskrit tri (‘three’), catur (‘four’) and panca (‘five’), and that the Sanskirt dua (‘two’) had been dropped in Eskayan bringing about a recalibration of the numeral sequence (Tirol 1993).

²⁰This presents another functional parallel to the Hangul system (created ca. 1443) wherein consonant characters were designed to represent human speech organs as a mnemonic to their place and manner of articulation.

²¹In this respect, the development of the Eskaya writing system has followed the same two-stage developmental sequence as the Kikakui (or Mende) script invented in Sierra Leone in the 1920s, wherein 42 semi-alphabetic characters were produced in a primary phase to be later followed by a further 153 purely syllabic characters (Dalby 1968).

²²Rizal was not the first to recommend reforms to Tagalog spelling. By his own admission he had taken inspiration from the prominent Filipino intellectuals Trinidad Pardo de Tavera and Pedro Serrano Laktaw who were already using elements of this new system. Indeed, the latter had gone so far as to revise the spelling of his own family name from ‘Lactao’ to ‘Laktaw’. He was not without his opponents. While his new orthography had the effect of clarifying the sound system of Tagalog, it also disguised and indigenized Spanish loanwords. Critics writing for the Catholic Review considered the foreign letter ‘k’ to be an unpatriotically ‘German’ imposition and an affront to mother Spain. One went so far as to sign an article with the provocative pseudonym hindí aleman (Tagalog: ‘not German’); this context is described in Thomas (2012: 153–166).