The Graphs

Variant forms of writing

Order of the graphs

Direction of writing

How to represent vowels

Three, five, or six vowels?

How to write ra,re-ri, and ro-ru

What to do with the final consonants of syllables

Features of Alibata (Usage Guide)

The Graphs

The script consists of three graphs which stand for vowels (or, according to some scholars, they represent glottal stops followed by vowels¹³), and fourteen graphs which represent syllables consisting of a consonant and the sound /a/.

FIGURE 3
Alibata

Variant forms of writing

The Spanish recorded ten variations of Alibata, all of which were almost entirely identical except for minor details. ¹⁴ Many of these differences are due to the features of the languages they were used to record. For example, the Ilocano version of the script lacks wa and ha, while the Pampangan version lacks ya,wa, and ha.

Order of the graphs

Scholars typically present the order of the graphs as conforming to the traditional order of other Indic scripts: a, e-i, o-u, ka, ga, nga, ta, da, na, pa, ba, ma, ya, la, wa, sa, ha. However, the Tagbunawa people have mnemonic rhymes which place the graphs in the following order: o-u, a, e-i, la, ma, da, ga, ta, na, ha, ba, sa, pa, ya, nga, wa. For example, one of the mnemonics is "U ai! Lamang daga ta nakabasa pauyat wawa," which translates to an insult and then the phrase, "If we cannot read, this is indeed shameful, for it is merely a child's game."¹⁵ Almost every segment of the rhyme is a value of one of the graphs. Still another order is exhibited by "The A B C's in the Tagalog language," an illustration of the Doctrina Christiana published in 1593 by Father Doming de Nieva using type devised by the Chinese convert Keng Yong. In this example, the graphs are ordered thus: a e-i o-u ha pa ka sa la ta na ba ma ga da ya nga wa.¹⁶ The Spanish eventually reordered the graphs to approximate the order of the Roman alphabet: a, e-i, o-u, ba, ca, da, ga, ha, la, ma, na, nga, pa, sa, ta, ya. The Spanish also transliterated the graphs into Roman letters so that it adhered to the conventions of written Spanish. For example was transliterated as ga, and was transliterated as go-gu, but was transliterated as gue-gui. Instead of ka, ke-ki, and ko-ku, the Spanish transliterated as ca, que-qui, co-cu. ¹⁷

Direction of writing

The consensus is that the script was written in columns from the bottom to the top, with each succeeding column written from left to right.¹⁸ A few Spanish missionaries claim the scripts were written in columns from top to bottom, with each succeeding column written from left to right. ¹⁹ Francisco Ignacio Alzina observed that people writing in the Samar-Leyte region wrote in boustrophedon, except it was from the bottom to the top for one column then from the top to the bottom for the next column, and so on, alternating. ²⁰ The orientation of writing is depicted in Figure 3 below. Apparently, the text could be read from the bottom to the top, or, after rotating the text ninety degrees clockwise, it could also be read left to right, like Roman letters. This is probably what the Spanish did, and this is the reason that graphs are now represented from the left to right instead of bottom to top.

FIGURE 4
Direction of writing

How to represent vowels

The vowel following the consonant can be modified by diacritical marks either above or below the graph. These marks are called kudlit. A kudlit placed above the graph changes the vowel to /e/ or /i/.

ba
be-bi

A kudlit below the letter changes the vowel to /o/ or /u/

ba
bo-bu

In 1620, Father Francisco Lopez introduced the cross kudlit in order to represent certain Spanish words less ambiguously in his Ilocano version of the catechetical Doctrina Christiana. ²¹ The cross kudlit is placed underneath the graph and functions much as the virama in the scripts of India, so that the graph represents only a consonant.

ba
b

Three, five, or six vowels?

This idea is suggested by the Roman transliterations of certain words, which sometimes vary according to region. For example , which means "woman", is written in Roman characters as "babae" in Tagalog and as "babai" in Ilocano.

How to write ra,re-ri, and ro-ru

In the earlier history of Alibata, ra, re-ri, and ro-ru were represented as , , (la, le-li, lo-lu.) But they could also be represented as , , (da, de-di, and do-du.) ²⁴ The interchangeability of "l" and "d" might be illustrated by the different words of different languages for the word that means "day." In Cebuano and Ilocano, the word used is aldaw. In Tagalog, the word is araw. But in all cases, it would probably be rendered as "a da u." In some Philippine languages, /l/ and /r/ are truly allophonic, that is, interchangeable. This is reflected in the fact that native speakers of such languages confuse the two when trying to speak languages where the two are distinct phonemes. In other Philippine languages, such as Mangyan, which still uses a form of Alibata, /l/ and /r/ are written with the same graph, but they are distinct phonemes. In still other Philippine languages, most notably Tagalog, /d/ and /r/ are allophones. The following table illustrates the cases in which /d/ is used, and the cases in which /r/ is used.

FIGURE 5
Occurrence of the allophones /d/ and /r/ in Tagalog

Case	Word	Hypothetical rendering	Syllabification
/d/ occurring at the beginning of a syllable at the beginning of a word	daan "road, street, throughfare"		da an (not ra an*)
/d/ occurring at the beginning of a syllable following a CVC syllable	dagdag "addition,extension"		dag dag (not dag rag*)
/d/ found in the end of a syllable	tadtad "chopped into pieces"		tad tad (not tar tar*)
/r/ occuring at the beginning of syllables in the middle of a word	nakaraan "bygone, past, former, departed" (derived from daan)		na ka ra an (not na ka da an)

The transformation of the /d/ to an /r/ is most apparent when changing an adjective into a verb. Normally in Tagalog, an adjective can be changed into a verb by adding the prefix "ma-." For example kita, which can mean "seen, can be seen, visible" can be transformed into makita, meaning "to be seen, to see." When this change is performed on an adjective that begins with a /d/, the /d/ becomes an /r/. For example dinig, which means "audible, can be heard." can be turned into marinig, which means "to hear, to be able to hear." As to the case where /r/ is the first letter of a word, this case was likely rarely if ever seen in pre-Hispanic Tagalog.

What to do with the final consonants of syllables

Like Linear B, the script used to render Mycenean Greek, the Philippine scripts only indicate the onset, or beginning, of syllables and do not display the coda, or the end, of a syllable.²⁵ Only CV syllables can be properly represented. The word maganda, meaning "beautiful," would be rendered as (ma ga da). The word sinta, meaning "love between man and woman," would be rendered as (si ta).

The convention is unclear about the final w of a syllable. In Tagbanuwa, it would be represented as a "u" but in Mangyan, it is not written. Thus, it is not certain whether Tagalog giliw (meaning "darling, a person much loved") would be rendered as or as

Not all codas are necessarily dropped when rendering speech in Philippine script. Some words such as masdan,(meaning "to look at or observe carefully") palabsin,(meaning "to let out, to put out") and tingnan,(meaning "to look at, to see") are actually contractions of masidan, palabasin, and tinginan, respectively,²⁶ so that they could be represented as (ma si da) instead of (ma da), (pa la ba si) instead of (pa la si), and (ti ngi na) instead of (ti na).

While the script cannot completely represent the Philippine languages, it is not an unsurmountable difficulty when reading it. As mentioned, the writing system of Linear B had the same problem when writing Mycenean Greek. A similar situation occurs in ancient Hebrew, which does not have symbols for vowels. The occurence of vowels were determined by context and through conventional usage. Also similar is the occurrence of homonyms in English, in which the meaning of a word such as "bear" which can either be an animal, or mean "to carry," must be determined through context. Hence, it is likely that an ancient user of Alibata could tell the difference between the love between a man and a woman (Tagalog sinta) and a type of string bean (Tagalog sitaw), both which could be rendered as

The incompleteness of the Philippine scripts are often attributed to the theory that these scripts were relatively new developments in the Philippine cultures, so that they did not have time to evolve more conventions to deal with the deficiencies. The coming of Arab culture and Islam through trade around the 9th century AD probably hampered this evolution and European colonization beginning in the 16th century completely disrupted this evolution. Moreover, since these scripts, in the strictest sense, were not an indigenous development, they were not particularly suited to represent Philippine languages.