Actes : La "découverte" des langues et des écritures d'Amérique

Fray Bernardo de Lugo: Two Sonnets in Muisca*

Nicholas Ostler

Linguacubun Ltd / London, UK

I. Overview of the Muisca Language

Muisca, also known as Chibcha, was spoken in the central altiplane of Colombia until the mid-eighteenth century. It is known from three missionary grammars (viz Lugo, 1619; Quijano Otero, 1883 and Lucena Salmoral, 1967, 1970; González, 1987) which also contain bilingual texts for pastoral use, and (in the case of the last) a Spanish-Muisca glossary. There is also at least one other (currently unpublished) Spanish-Muisca glossary (Vocabulario Mosco 1612).

Ostler 1992a exhibits the main features of Muisca syntax, which is a strictly SOV. Ostler 1992b explores Muisca vocabulary, and Ostler 1993 discusses the closely interrelated set of case-particles, postpositions and subordinating conjunctions.

In what follows, I give sources for all Muisca phrases quoted. These are in the form of initials (L - Lugo, GP - González de Pérez, LS - Lucena Salmoral, V - Vocabulario Mosco) and a page-number. These page numbers refer to the pages of the published editions, and not to the original pages of the manuscripts. In the case of Lugo and the Vocabulario Mosco, the editions are facsimiles of the originals, so I there give the original page numbers followed by r - recto and v - verso, since leaves, rather than sides, are numbered there.

II. Historical Context of Lugo's Grammar and the Spanish Sonnet

The city of Santa Fé de Bogotá was founded in 1538, and a number of churchmen, starting with Gonzalo Bermúdez, took up the analysis and teaching of Muisca from the very earliest days of Spanish settlement. González (1980:76-80) lists twelve works which appear to pre-date Lugo's grammar, but none is extant.

Bernardo de Lugo was born in the New Kingdom of Granada late in the sixteenth century. He became a Dominican friar, and in 1615 he was appointed magister linguae indorum, after which he began preaching in Muisca. He was appointed to the chair of Muisca in Colegio del Rosario, the oldest college in Bogotá. On 1 August 1617 he received a commission from Friar Gabriel Giménez (given in full in Ortega, 1978: 64) to write an Arte (i.e. Grammar) and Confessional in Muisca. He worked fast, and received official approval of his work from the three official examiners on 19 and 20 February 1618. It was then sent to Madrid for printing, and was issued by the house of Bernardino de Guzman in 1619. It appears that Lugo was unable to correct the proofs, and the work came with such a strict royal privilege -- no reprints for 10 years, under pain of a fine of 50,000 maravedis -- that no correction of the many misprints was possible. There was to have been an accompanying dictionary and catechism, but this seems never to have been printed (Ortega, 1978: 65).

A touching feature of the work is that it begins with three sonnets, one in Spanish, and then two in Muisca. The authorship is unclear, but they are so fulsome in their praises of Lugo that one can only hope that Ortega is right in attributing them to his "compañeros", and not the author himself. The Spanish sonnet (spelling modernized) runs as follows, with an English translation:

SONETO.	SONNET

Quién eres tu que tan ligera vuelas?	Who are you who fly so light?
La lengua Chibcha soy. Ado caminas?	I am the Chibcha tongue. Where are you bound?
Del nuevo Reyno, a tierras peregrinas,	From the new Kingdom, to exotic lands,
Que tendrán mis verdades por novelas.	Which will take my truths for new ones.

Dices muy bien que a todos nos desuelas	You well say that you distress us all
Con tu profundidad, dí qué imaginas?	With your depth; say, what do you imagine?
Que estudiando sabrás lo que adivinas,	That by studying you will know what you intuit,
Que el docto Lugo preside en mis escuelas.	That the learned Lugo is foremost in my schools.

Pusome en Arte siendo yo intricada.	He put me in a Grammar, since I am intricate.
Y de chontal me hizo tan ladina,	And from a rude thing he made me so urbane
Que causo admiración al mundo todo.	That I cause admiration throughout the world.

Por él pienso quedar eternizada,	Thanks to him I expect to remain immortalized,
Y su opinión de hoy mas será divina,	And his repute from today on will be godlike,
Que él sólo alcanza mi substancia y modo.	In that he alone attains my substance and mode.

The themes which run through the Muisca sonnets are all present here: the intrinsic difficulty of Muisca, Lugo's unique understanding of it, and the resulting mead of praise that he can expect.

III. The Muisca Sonnets: Textual Emendation, Analysis and Translation

The sonnets in Muisca occur immediately after the Spanish sonnet, after the title page but before all the other front-matter including errata. The errata do not mention any passages in the sonnets.

Each line is given first as it appears printed in the codex (line-breaks are replaced by [), then with any emendations which I propose. Next, I transcribe the text according to the spelling conventions established later in the seventeenth century and used in all other Muisca texts. (Discussion of most of the points at issue here can be found in Constenla 1984 - but note comments on line 1.9.) I then give a morphological analysis of the text, with glosses for each morph, and finally my translation into English. After this there is discussion of any points arising.

Sonnet 1:

1.1

MVγÅca micâta cubun cħoqγ vca- [ Åûca,

MVγÅca micâta cubun cħoqγ vcaÅûca,

muysca micata cubun choc ucasuca

muysca micata cubun cho-c uca-suca

Muisca beautiful language good-Adv. know-Imperf. Participle

One who learns well the beautiful Muisca language,

In the sonnets, Muysca and chibcha are used apparently without distinction to refer to the language. (1.1, 2.2; 2.1, 2.9) Muysca is also used to refer to people generally (1.13), and once perhaps (2.12) more specifically to refer to indigenous people. micata is glossed "bella cosa, id est mala" ("fine, i.e. bad") at V 9r.

The sense requires that the verb forms ending in -suca in 1.1, 1.5, 2.1, 2.13 be analyzed as participles. GP 85, L 51v, LS 63 all state that -suca can have this function. Nowhere else, however, have I found the root uca without the stative suffix -n-. when followed by -suca. I have inferred that it can have non-stative meaning "come to know", "learn": the future participle is seen in 1.3.

1.2

Hiaqûnhân cħicħiranγnga, hocâbga- [ nân,

Hiaqγ mhân cħicħiranγnga, hoc âbganân,

iahac mahan chichiganynga, hoc abganan,

ia-hac	maha-n	chichigan-ynga,	ho-c	a-b-ga-nan,
now-how	2sg-Locat.	strange/rare-Future	3rd-Dative	3rd-Trans.-teach-if

if he should teach how it is strange to you

The common adverb iahac Lugo writes as hiaq: e.g. L 142v = 125r: Hia quaguèn umqyzhân (iahac aguen umquyzan) = "Porque no lo has hecho?" ("why have you not done it?"). But the rest of Hiaqûnhân is obscure. All the other examples of post-consonantal h (outside the ligatures öh and ch) seem to represent a VhV vocalism in the established spelling (e.g. nhocân 1.12, xhicâ 1.13, abhâqγnöhînga 1.14). Stella González de Pérez suggests that the û may be a mistake for γ. If we suppose the m of mhân to have been corrupted, we get the text suggested here. I have replaced the r of cħicħiranγnga with a g, though most of Lugo's r's correspond to standard ch (cf. γerqγ 2.3 below), since r could be close phonetically to an intervocalic [g], and makes good sense.

V 36v: "Rara cosa, id est, qual y qual entre muchas" = chichygansa [o] chichegansuca.

The 3rd person pronoun hoc governed by bgasqua "teach" (cf. GP 256, LS 55) appears gratuitous, but may be necessary to fix the sense of the bgasqua which can also mean "make, render" with a different construction (GP 269). It is natural to assume that this pronoun hoc marks the human recipient, in this case an indefinite 3rd person. But this clashes with the 2nd-person pronoun we have reconstructed within Hiaqûnhân.

1.3

Libros, cubun γn vcanγnga, ab chihi-qγnân,

Libros, cubun γn vcanγnga, abchihiqγnân,

libros, cubun yn ucanynga, abchihiquynan,

libros,	cubun	yn	uca-nynga,	a-b-chihi-quy-nan,
books	language	Loc-Prefix	know-Future	3rd-Trans.-write-Past-if

if he should write books, to explain the language,

The yn prefix, seen also at 1.9, is documented in L 106r-107v and V 2v. This proclitic, found only in relative clauses, marks the antecedent (libros) as in a locative relation to the clause: i.e. books, in which (the indefinite subject) will come to know the language. Cf. Ostler, 1993:IIIC on this construction.

1.4

Apuγqγ cħiê öhaguenγ γqγ anguÅûca

apuyquychie z aguene yc angusuca

a-puyquy-chie-z	a-gue-ne	y-c	a-n-gu-suca
3rd-heart-light-Epenthet.	3rd-be-Stat.	3rd-Allative	3rd-Pass.-say-Imperf.

he will be told that he has ability

puyquy-chie is a common word for "ability". chie, "the moon", also means "light" in general, (V 28r); it also has the metaphorical sense of "honour" in 1.10, 2.10 and 2.15 (GP 271, V 25v). For the sense, a future would be expected here: perhaps Lugo takes the licence of using the present for the rhyme.

This is the first of four problematic -suca terminations (cf. 2.1, (2.10) and 2.13). They are problematic because they are formed from verbs of the -squa conjugation, viz gusqua "say" and bquysqua "do". Adelaar (this volume) draws attention to GP 332 where a contrast is made between -husqua "venir" (come) and -husuca "venir a menudo" ("come often"). This is evidence that some verbs at least could take both -squa and -suca terminations.

1.5

γngâ xis chicubun γqγ vmöhanÅûca

nga sis chicubun yc umzansuca

nga	sis	chi-cubun	y-c	um-za-n-suca
and	this	1pl-language	3rd-Allative	2sg-place-Stat-Imperf. Participle

and as one who is set on this our language

The meaning, and the morphology, of um-za-n-suca are not clear. It is here analyzed as a derived form related to bzasqua "put, place" exactly as quynsuca "happen, meet, fix" (V 1v, 20v, 4r) is related to bquysqua "do". GP 125-6 give a list of "verbos finitivos" which are passive statives with the -n- suffix, but this za-n- is not among them. The grammars give no general rule for deriving -suca forms from statives in -n- but a-quyn-suca is widely attested (e.g. LS 212: ipquabie-z ys aquynsuca something happens to them).

1.6

Vmpuγqγ chieöhγ hγÅquî γnöhaÅasân

umpuyquychie z hysqui nzasasan

um-puyquy-chie-z	hysqui	nza-sasan
2sg-heart-light-Epenthetic	so	Neg. Copula-if

if your ability is not as great

A peculiarity of Lugo's account of Muisca grammar is that the conditional suffix with the negative copula is -sasân not -san. (L 29r cf. GP 80, LS 37).

1.7

Muγyas agaöhînga, γnga γÅqhicħan

muyias agazinga, nga ysquihichan

muyia-s	a-ga-zi-nga,	nga	y-s-quihicha-n
clear-Perlat.	3rd-become-Neg.-Fut.	and	3rd-Perlat.-because (lit. foot-Loc.)

it will not become clear, and so

The perlative -s is superficially odd governed by gasqua "become": it usually requires allative -c. However, muyias is a general adverbial form of muyia "clear" (e.g. V 1v, 17v, 12v, 16v).

1.8

ögaqγ ÅuâÅâ mabiê γqγ vmÅunÅûcà.

öhaqγ ÅuâÅâ mabiê γqγ vmÅunÅûcà.

zac suasa mabie yc umsunsuca

za-c	sua-sa	mabie	y-c	um-sun-suca
night-Allat.	day-Perlat.	much	3rd-Allat.	2sg.-think-Imperf.

you ponder about it long by night and day.

ög is an impossible ligature, since ö is always followed by h.

mabièis quoted in L 5v as meaning "much, many". It is clearly derived from fie, the normal word in GP (esp. 146): b and f do not contrast in Muisca (cf. Constenla, 1984 § 3.1.2). The first two words seem to be a fixed phrase with different cases for za and sua : V 17r "De noche y de dia" = Zaque suasa. mabie cannot qualify them, since if it did it would bear the case-marking: e.g. GP 323: muysca cuhuma-s "as a powerful man".

1.9

Arte γnchicħicħua nγnga yê vmqγ nân,

Arte γnchicħicħuanγnga yê vmqγnân,

arte yn chichichuanynga ie umquynan,

arte	yn	chi-chichua-nynga	ie	um-quy-nan,
grammar	Loc. Prefix	1pl-learn-Future Participle	road	2sg-make-if

If you should make a way for us to learn the Arte

Lugo's y before a vowel corresponds to what was later written with i. (cf. anyanua? Anyanga. 2.11.) He could use y for this allomorph, since he had γ for the distinct vowel later written y.

Note that Lugo's orthography distinguishes the ch of chi- "we", chie "light, glory" from the barred cħ of cħo "good", cħicħua "learn"and cħibcħa. This distinction was lost in the later orthography. Although Constenla 1984 § 3.1.5.1 remarks that Lugo "tended to write the ch in Muisca with a bar on the h", he does not note this apparently systematic distinction in Lugo's phonology.

The verb bchichua-suca means "to learn" (V 6v, 8v). It lacks its -b- here as it is participial (GP 86).

1.10

Ipqua bhoöa noabê vmcħiê chibqγngâ

Ipqua bhoöha noabê vmcħiê chibqγngâ

ipqua bohoza nua be umchie chibquynga

ipqua	bohoza	n -ua-be	um-chie	chi-b-quy-nga
what	with	Fut.-Interrog.-Deliberat.	2sg-glory	1pl-Trans.-make-Future

with what shall we sing your praises?

I have conjectured nua , the future interrogative of the copula, which makes good sense. But Lugo himself (L 30r) gives the form as nuâ not noa.

Ostler 1992a: §6 describes the fronting of question-word and Interrogative Copula (ua). Usually the main-verb is participialized in this construction: but not (as here) when the questioned element is a postpositional phrase. The particle be is seen in many questions (e.g. GP 306-309 passim), but usually attached to the main verb, not as here to the copula.

The sense seems to be: "what materials can we find to construct your glory?", suggesting that everything here is too lowly.

1.11

Con MuγÅca atabè vmqγ yê amiöhin- [ ga.

Con MuγÅca atabè vmqγ yê amiöhinga.

con muysca atabe umquy ie amizinga

con	muysca	atabe	umquy	ie	a-mi-zi-nga
cry	Muisca	any	2sg-make Past	road	3rd-pass-Neg.-Fut.

You have made a way that no Muisca cry will pass

This is a difficult line. con is the usual word for "weeping", common as a verb-stem but also attested as a noun: V 28r: "Llanto " = cone. (with the obligatory e or y added to every final n). Adelaar (p.c.) notes that in Quechua waqay can mean "sing" as well as "weep", but such a sense for con- is not noted in any of the Muisca vocabularies. Lugo seems to have allowed himself some latitude in word-order, placing the main verb in the middle of the line, and spreading the relative clause qualifying ie all round. (This spreading of relative clause constituents is quite common: cf. Ostler, 1992a : §2.8.) amizinga needs then to be taken as a participial: none of the grammars explicitly give forms for negative participials, and the prefix a- is problematic (cf. 2.1 and 2.13); there is another negative participial (formed differently) at 2.7.

The result of this licence is that the line is quite a precise phonic echo of line 1.10 which precedes.

1.12

γnga xis qγcacâ vmguγ nhocân;

nga sis quycaca umguy nohocan

nga	sis	quyca-ca	um-guy	nohocan
and	this	land-Inessive	2sg-be	although

and even though you are in this land

Constenla 1984 : 86 - § 3.1.7 -notes correctly that x and s are in free variation before i, and probably pronounced [Å]. Cf. xhicâ in the next line.

1.13

MvγÅcà vmguγ γpqvanâ xhicâ gungâ,

muysca umguy ypquana sihic agunga

muysca	um-guy	ypqua-na	sihic	a-gu-nga
person	2sg-be	after-Loc.	thus	3rd-say-Future

people after you have been will speak thus:

I have transferred the a from the end of xhicâ onto the beginning of gungâ, for verbal concord. The a at the end of xhicâ would have been optional anyway.

1.14

Vmhγca Bernardo vqûqγ abhâqγnöhînga.

Vmhγca Bernardo vbûqγ abhâqγnöhînga.

umhyca Bernardo ubuc abahaquynzinga

um-hyca	Bernardo	ubuc	a-bahaquy-n-zi-nga
2sg-name	Bernardo	ever	3rd-abase-Stat.-Neg-Future

Your name, Bernardo, shall never be brought low.

There is no known word vqûqγ (ucuc): hence my emendation. Cf. GP 284: "Nunca jamás." : Hataca [o] ubuca …

The last word is unattested elsewhere. However, V 9r has: "Baxa cosa, vil " = vahaque muysca [o] bahaque guacha guasgua. bahaque clearly means "low, vile".

Sonnet 2:

2.1

Xiêgua cħibcħa cubun mabie agusû- [ ca? [ Bernardo guγ.

Xiêgua cħibcħa cubun mabie agusûca? Bernardo guγ.

xie gua chibcha cubun mabie agusuca? Bernardo gue.

xie-gua	chibcha	cubun	mabie	a-gu-suca?	Bernardo-gue.
who - Interrogative Copula	Chibcha	language	many	3rd-say-?Imperf. Participle	Bernardo - Affirmative Copula

Who is it who describes at length the Chibcha language? Bernardo.

The problem here is the apparent present participle of gusqua, "say", whose 3rd person present participle is by rights guisca (cf. GP 230 with inflexion paradigm on GP 82). Yet the sense seems clear, and a participle is required in a question beginning xiegua (cf Ostler, 1992: §6).

As a -suca verb, agusuca's participial ending is standard: the prefix a-, though is obscure.

2.2

MuγÅca hγÅqui vcasûca aguenuâ? [ Aguêöhâ.

MuγÅca hγÅqui vcasûca aguenuâ? Aguêöhâ.

muysca hysqui ucasuca aguenua? agueza.

muysca	hysqui	uca-suca	a-gue-n-ua?	agueza.
Muisca	so	know-Imperf. Participle	3rd-be-Stative-Interrog. Copula	3rd-be-Negative

Is there anyone who knows Muisca as well? There is not.

2.3

γngà γerqγ atabe amuγs apqua öhâ- [ guâ? [ Apquâöhâ.

γngà γerqγ atabe amuγs apqua öhâguâ? Apquâöhâ.

nga yechica atabe amuys apquazagua? apquaza.

nga	yechica	atabe	a-muys	a-pqua-za-gua?	a-pqua-za
and	other	anyone	3rd-toward	3rd-reach-Neg.-Interrog. Copula	3rd-reach-Neg.

And does no-one else attain it? None does.

For the ch corresponding to Lugo's r before [k], cf. L 89r, where huirqγ, huircâ are quoted among the "preposiciones", corresponding I presume to V 43r (ze-)huichquy : ""ad sensum vel notitiam meam" (in my sight). González de Pérez (p.c.) also points out the variants for "old" tybara, [o] tybacha (GP 333).

yechica is quoted directly at GP 157: ("es lo mismo que alius …").

The verb sense, if strained in English, parallels the Spanish sonnet's "Que el solo alcança mi substancia y modo."

2.4

Ocanxînga hγÅqγ quî amucanγ? Anguaguêguγ.

ocanxinga hysquy amucane? anguaguegue.

ocanxinga	hysquy	a-m-uca-ne?	anguague-gue.
truly	so	3rd-Trans.-know-Stat.	awe-Affirmative Copula

Does he truly know it so well? It is awesome.

I know of no word quî outside these sonnets. I conjecture it has slipped in after the -qγ of hγÅqγ.

ocanxinga, which occurs again in 2.9, is a an idiom, meaning "certainly, truly" (GP 160, V 15r). anguague is attested in V as a translation for various terms of disgust and fear: "terrible", "espantable", "abominable". However, V 40r says that it can be a synonym for much: "y asi suelen decir: guasga z anguague, id est, ynapuyquyne" which would mean "there are a lot of children". Here it expresses exaggerated admiration for Lugo's knowledge.

2.5

Anguaqγ aguen qhicħan öhγb xieÅγ.

Anguaguêqγ aguen qhicħan öhγb xieÅγ.

anguaguec aguen quihichan zebsiesy

anguague-c	a-gue-n	quihicha-n	ze-b-siesy
awe-Allative	3rd-be-Stative	because (lit. foot-Loc.)	1sg-Trans.-fear

Because it is awesome, I am afraid

I have inserted an extra guê in Anguaqγ for the sense. I have only found angua itself in the phrase angua bquysqua "hear a noise" (e.g. GP 318, V 37v).

2.6

γs gues qγ atabe öhγbnγ mucan öha

γs guesqγ atabe öhγbnγ mucanöha

ys guespqua atabe ze[bny] mucanza

y-s	guespqua	atabe	ze-[bny]-m-uca-n-za
3rd-Perlative	like	anyone	1sg-[?]-Trans.-know-Stative-Neg.

I do not know anyone like him

It is difficult to make any sense of the syllable bny which looks like the stem of the verb mny-squa "give", "take" or "put". Unfortunately, it seems to be necessary to make the syllable-count in 2.6 equal to its balancing line in the sonnet, 2.7.

2.7

γnga cubun etaqγn öhγ mvcan öhâ

γnga cubun etaqγn öhγ mvcanöhâ

nga cubun etaquynz mucanza

nga	cubun	etaquyn-z	m-uca-n-za
and	language	nothing-Epenthetic	Trans.-know-Stat.-Neg. (-Participle?)

and one who knows nothing of the language

öhγ in this line, and the next, have been interpreted as "adornos", i.e. "epenthetic". LS 56 states: "el adorno de la z se pone despues de la n quando inmediatamente se sigue vocal (maxime a)." i.e

.ø -> z / n_# V. But one does not have to read far to see that the adornos are not so narrowly restricted e.g. LS 210 Muysca atebie-z muhuc aguahaica-n " some man's harm to you". And cf. line 2.11 below. By contrast, if these öhγ are interpreted as 1sg prefixes, the translation becomes slightly more difficult: "I who know nothing… can I know very well his language?"

V 30v, under "Ninguno", has etaquyn magueza, literally "at base non-existent".

2.8

γs acubun cħoin öhγ mucanbê.

ys acubun choinz mucanbe?

y-s	a-cubun	cho-in-z	m-uca-n-be?
3rd-Perlative	3rd-language	good-Superlative-Epenthetic	Trans.-know-Stat.-Deliberative

can he know very well his language?

2.9

Arte cubun Cħibcħa o cas abqγ? Ocan xinga.

Arte cubun Cħibcħa ocas abqγ? Ocanxinga.

arte cubun chibcha ocas abquy? ocanxinga.

arte	cubun	chibcha	oca-s	a-b-quy?	ocanxinga.
grammar	language	Chibcha	= a-uca-s: 3rd-know-Gerund	3rd-Trans.-do+Past	truly

Did he know the Chibcha language when he made the Arte? Truly.

i.e. "knowing the Chibcha language, did he make the Arte ?" L 121v-122v gives an account of the -s which attaches to all but the last of a series of parallel verbs.

2.10

γs qħihicħan achie öhγ anguÅungâ? AnguÅungâ.

γs qhicħan achie öhγ anguÅucâ? AnguÅucâ.

ysquihichan achie z angusuca? angusuca.

y-s-quihicha-n	a-chie-z	a-n-gu-suca ?	angusuca.
3rd-Perlat.-because (lit. 3rd-Perlat.-foot-Loc.)	3rd-glory-Epenthetic	3rd-Pass.-say-Imperf.	ditto

And so will his glory be spoken? It will be.

I have emended both in question and answer. As the poem stands, line 2.13 lacks any rhyme. Moreover, angusunga , if accepted, would have contained an extra morph -su- which is not readily explainable. angusuca is also dubious, since the passive should be angusqua. But it is not the only illegitimate -suca form: cf. 1.4, 2.1 and 2.13.

2.11

Acħicaöhγ mabie an ya nua? An yanga.

Acħicaöhγ mabie anya nua? Anyanga.

achica z mabie anianua? anianga.

achica z	mabie	a-n-ia-n-ua?	a-n-ia-nga.
3rd-cause (lit. 3rd-top) - Epenth.	much	3rd-Pass.-open-Fut.-Interrog.	3rd-Pass.-gather-Fut.

For this will much be revealed? It will be.

L 34r gives chica-na/- ca meaning "por ocasión". In GP 114 the passive of ze-b-ia-squa is given as cha-n-na-squa, not *cha-n-ia-squa, which would correspond to what we have here. Perhaps this is an earlier or dialectal form, or perhaps the -nn- is used to represent a nasal palatal like n~. At any rate, Lugo's spelling a-nya- makes it quite clear that the i in the stem is not syllabic. And the realization of the transitive prefix as -b- confirms this: before a syllabic vowel, it would appear as -m-, as it does, e.g. in ze-m-imy-suca. "change".

2.12

HγÅpqua MuγÅca atabê öhγmhixtγö- [ hinga,

HγÅpqua MuγÅca atabê öhγmhixtγöhinga,

hyspqua muysca atabe zemistyzinga

hyspqua	muysca	atabe	ze-misty-zi-nga
such	person	any-Epenth.	1sg-find-Neg.-Fut.

I shall never find any such person/Indian

GP 157: "Yspqua significa talis [o] exeogenere, como cha-yspqua-gue, tal soi…"

2.13

Aquγca hγÅqui cħo γqγ abqγsûca

aquyca hysqui cho yc abquysuca

a-quyca	hysqui	cho	yc	a-b-quy-suca
3rd-land	such	good	3rd-Allative	3rd-Trans.-do-Imperf. Particp.

who does such good for his land

abquysuca is another non-standard participle, which should be by the rules quisca (L 41v, GP 83) -cf. 2.1 above. Furthermore, by the case-valences set out in LS 53-56 yc bquysqua should mean "seize him": "do him good", the apparent sense here, should be hoc cho bquysqua. The construction appears very similar to that which gave trouble in 1.11, namely a main verb surrounded by various parts of the long relative clause which is its direct object.

2.14

Achiê, a honra, mabiê γqγ angumgâ.

Achiê, a honra, mabiê γqγ angungâ.

achie, ahonra, mabie yc angunga.

a-chie,	a-honra,	mabie	y-c	a-n-gu-nga.
3rd-glory	3rd-honour	many	3rd-Allat.	3rd-Pass.-say-Future

His glory, his honour, will be told to many.

IV. Metrics and Rhyme-Scheme

The Spanish sonnet's metre is iambic, with five main stresses, and feminine rhymes throughout. It is more difficult to discern any rhythmical pattern in the Muisca sonnets, especially since nothing is known about stress, accent or vowel-length. The lines have from 11 to 14 syllables in the first of them, and from 11 to 17 in the second, but there are doubts about the scansion of diphthongs with u, i and h .

The rhyme-pattern in each of the three sonnets is different. They are:

Spanish: ABBA ABBA CDECDE

Muisca 1: ABBA ABBA BCCBCC

Muisca 2: ABBA ABBA' CDCCDC

Muisca 2 is noticably the roughest of the three, with a poor rhyme of line 2.8 (bê) to line 2.5 (sy), and line 2.13 failing to rhyme with anything unless line 2.11 is emended.

V. Conclusion

The sonnets in Muisca are so extravagant in their praise of Lugo that to modern ears they sound insincere, rather like the work of a friendly colleague teasing him about his aspirations. The only evidence about the author(s) is chicubun "our language" in 1.5. Could this imply that the writer was a native speaker of Muisca? This might clarify some points where the writer's morphology and syntax does not conform with Lugo's own prescriptions.

But the sonnets may have been intended more to encourage buyers than to flatter the author. Fray Domingo de Santo Thomas's 1560 grammar and dictionary of Quechua begin with Latin elegiac verses praising the author's intellectual feat, but also urging the reader to buy the books. Nevertheless, such verses do seem to have been rare: in the Conde de la Viñaza's Spanish bibliography on American indigenous languages no other poetic invocations figure among the 244 titles he reviews up to the end of the seventeenth century.

The sonnets may also be seen as the fruit of a teaching tradition which emphasized imitation of traditional authors: thus the ponderous question-and-answer style of the second Muisca sonnet may be a conscious echo of the Catechism: its heaviness certainly contrasts with the much more elegant dialogue of poet and language in the Spanish sonnet.

Whatever their purpose or authorship, it is fortunate that they were printed. For these twenty-eight lines are now the full extent of original literature known to have survived in Chibcha.

References

Adelaar, Willem

1993 "Les catégories verbales "conjugaison" et "genre" dans les grammaires de la langue chibcha". This vol.

Constenla Umaña, Adolfo

1984 "Los fonemas de muisca". Estudios de linguistica chibcha, ed. Enrique Margery P. & Adolfo Constenla U. Depto de Linguística, Universidad de Costa Rica .

González de Pérez, María Stella

1980 Trayectoria de los estudios sobre la lengua chibcha o muisca. Bogotá: Instituto Caro y Cuervo.

1987 'Diccionario y Gramática Chibcha' - MS anónimo de la Biblioteca Nacional de Colombia. Bogotá: Instituto Caro y Cuervo. [GP]

Lucena Salmoral, Manuel

1967 "Gramática Chibcha del siglo XVII". Revista Colombiana de Antropología XIII. 31-90. Continued ibid. XIV. 201-220. [LS]

Lugo, Fray Bernardo de

1619 Gramática en la lengua general del Nuevo Reyno, llamado mosca. Madrid: Bernardino de Guzmán. (Facsimiles: (1) "Resurrección de una lengua" pref. M. Alvar, Ediciones Cultura Hisp.del Centro Iberoamer. de cooperación, Madrid, 1978; (2) Colegio del Rosario, Bogotá, 1981.) [L]

Ortega Ricaurte, Carmen

1978 Los estudios sobre lenguas indígenas de Colombia. Bogotá: Instituto Caro y Cuervo.

Ostler, Nicholas

1992a "Syntactic Typology of Muisca - a Sketch". Language in the Andes. Univ. Delaware Press, Cole, Hermon & Martin ed.

1992b "Estudios Lingüísticos sobre los Textos Litúrgicos en Idioma Muisca". presented at 6th Congress of Anthropology in Colombia, Bogotá.

1993 "Case-Marking and Complementation in Chibcha". 13th International Congress of Anthropological and Ethnological Sciences. Mexico City, 2 August.

Quijano Otero, José María

1883 "Gramática, frases, oraciones, cathezismo, confessonario y bocabulario e la lengua chibcha 1620…" Congreso internacional de americanistas 4., Madrid 1881: 226-295. Madrid: Imprenta de Fortanet.

Viñaza, El Conde de la

1892 Bibliografía Española de Lenguas Indígenas de América. Madrid (facsimile : Madrid: Ediciones Atlas 1977).

Vocabulario Mosco. 1612. MS 2923 en la Biblioteca del Palacio Madrid. [V]

* I wish to thank the President of Colegio del Rosario, Santafé de Bogotá, for his kind gift in 1989 of a facsimile of Lugo's grammar, without which this work would have been impossible. Also Stella González de Pérez for her discussion of an earlier version of these translations. The audience at the Colloquium, especially Gerald Taylor and André Cauty, enriched my understanding of the wider context, and provided many of the points made in section V. All errors remain my responsibility.