The Madonna and the Horse
Becoming Colonial in New Spain and Peru
TOM CUMM I NS

AMONG THE MARVELS praised by Europeans in the New World, in New Spain
(Mexico) and Peru in particular, was the artistic ability of its native inhabitants.
They not only made marvelous things for themselves before the Spanish
invasion ,' but also, in less than a generation, were able to produce Europeanstyle images as pleasing to the eye as anything created in Europe. In fact, the retraining of native artists to create such images was interpreted as a sign of the
success of the Spanish mission.
Thus one finds the inveterate conquistador, Bernal Diaz de! Castillo, writing
in a chapter entitled ' How We Instructed The Indians Of New Spain In The Holy
Doctrine ... And How We Taught Them The Crafts Used In Spain ... ' that 'before
this could be accomplished the Spaniards first had to extirpate the idolatry of the
Indians .'' He then goes on to state that after this success native artists learned to
paint the Passion of Christ so well that if one did not see them do it, it would be
impossible to believe that a native had made such images; and these in his
judgement made the work of three Mexican artists, Andres de Aqu ino, Juan de
la Cruz, and el Crespillo, comparable in stature to that of the ancient Apelles,
the modem Berruguete, Michelangelo, and the most recent master from Burgos.
High praise for their craft sure ly, but there is more to Diaz's words than mere
praise. The transformation of the makers of idols for idolatry into makers of
images for Christian contemplation was an essential step in the transformative
acculturation of native Americans. In the eyes of sixteenth-century Spaniards ,
you became what you made.
One can see the ideology of this transformation perhaps no more clearly than
in the different representations of artists in one of the great colonial Mexican
manuscripts, the Florentine Codex produced by the Franciscan Bernardino de
Sahagun with the aid of Aztec artists and scribes trained by him. In one section
where an Aztec sculpto r in wood is depicted making idols for the ancient
religion, the image is composed of just Mexican elements set against a flat ,
empty background (figure 1). In another section where the ability of native
artists is addressed (rather than what they once produced), their representation
52

THE MAD O N N A AN D THE

HORSE

Figure r. Construction of an idol, Florentine Codex, Book I, append ix,
folio 26. From Benardino de Sahagun, Historia general de las cosas de Nueva Espana,
Cadice Florentino (Flo rence and Mexico, 1979), I.

53

N AT I VE ART I ST S AND PATRO N S IN CO LO N I AL LAT I N A
M
ERI

CA

Figure 2. Gathering nacazcolotl , dyeing with it, writing with it, Florentine Codex,
Book XI, folio 218 verso. From Sahagun, Hi storia general ( 1979 ) , 111.
is based on European woodblock prints and set within an illusionistic background (figures 2 and 3). Here, the depiction of native craftsmen has been infusedwith
a sense of status commensurable with their European counterparts; they have
been distanced from their idolatrous connections through the use of a
European prototype.'
The status and function of native colonial artists and what they produc ed,
however , was defined in cases like these first through writing. That is, these
examples all occur within a European context that ultimat ely gives precedence
to the written word, and we tend to lose sight of the fact that such techniques of
54

THE MAD O N NA AND T H E HORSE

Figure 3. Jost Amman , Der Brieffinaler, woodcut , in Harmann Schopper ,
Panoplion (Frankfourt , 1568).

inscription are a part of the complex strategy of conquest.' In this paper I want
to move away from such self-conscious colonial works in which the European
symbolic technology of writing ultimately dominates. Writing is presumed too
often by modern scholars to control the epistemological discourse of'the other;'
and, since they study mostly writings by Europeans, what is revealed is only
what Europeans thought , knew, or loved.' There is thus only one 'other:' the nonEuropean. Natives remain captured in the written word as objects of European
self-reflection. • To move beyond the bonds of writing, I wish to turn to native
forms of symbolic technology, visual images that frame and record knowledge
55

N AT I VE ARTISTS AND PATRONS IN CO LO N I AL LAT I N A M E R IC A

within the native world. Many forms produced by native artists in the colonial
period in and of themselves were not new or unusual to native expression
before the Conquest, but they were fundamentally altered in meaning because
they were produced after the Conquest, bringing up issues that can arise only
from a European experience and a colonial context.
In contrast, I will concentrate on two examples of European forms found in
traditional contexts - one, an image found in a 1531 manuscript from Mexico,
and the other, an image found on a gold embossed band from Peru (figures 4 and 5)
- that I will treat as case studies in which we cannot ascribe the shifts and
disjunctions in visual forms to purposeful European agendas. Here, native
artists themselves have taken up certain European pictorial conventions to communicate about things which are 'new' to them and must be addressed because
they have impinged directly on their lives. I am not so concerned with Aztec or
Inca constructions of the European as ' the other ' as much as with the internal
native dialogue that comes about because of the European 'other.' This distinction
is important because, as a concept, the ' New World' is more a matter of time than
place; it is a time in which two ancient world cultures collided through the aggressive act of invasion, to begin a process of continuing dialectical permutations
where native participation was neither as passive, reactive, or silent as is so often
believed. In addition, colonial art is not a single entity but multiple phenomena,
and native artists were from the very beginning capable of producing works that
expressed a colonial content in their own terms and for their own needs. Native
artists therefore cannot simply be enclosed between the poles implied by the
terms 'makers of idols' and 'imitators of Apelles or Michelangelo .'
Rather , in the very early colonial period before the so-called 'extinction' of preColumbian motifs,' the crisis provoked by the European invasion involved not
only destruction but also dynamic change in native forms of representation. In the
case of the Mexican document the motif in question is a representation of the
Virgin and Child and in the Peruvian gold piece it is a man riding a horse. These
new images are placed within a context of traditional native forms, however , and,
although as European-style images they may appear intrusive and awkward, they
operate as part of native expression. Their deployment is not as much about contestation between two cultures, as it is about the taking-on of new forms and their
becoming something different . In both examples , the native artist 's ability to
improvise comes in response to a need to renegotiate formal conventions within
new systems of power in the colonial world without abdicating the capacity of
traditional forms to signify in that world as well.' Nonetheless, the result of artistic
improvisation is not simply the inclusion of new formal and iconographic
elements , it is about what the new images mean in the native world.

THE MADONNA AND THE HORSE

Figure 4. Painting Number 5, Huejotzingo Codex, 1531.
Harkness Collection, Library of Congress.

Figure 5. Sixteenth-century gold band from South Coast of Peru, 10 x 2 7 / 16 inches.
The Art Institute of Chicago, number 55.2608.
57

N ATI

VE

ART

I ST

S

AND
AM

PATRONS
ERI C
A

IN

CO

LO N I AL

LATIN

Figure 6. The Spaniards making boats for the assault on Tenochtitlan.
From Diego Duran, Historia de las indias de Nueva Espaiia, chapter LXXVII, about 1579-81.
It can be argued that there is an internal dialogue within such 'images about
images' that references and juxtaposes the two worlds. Moreover , the dialogue
generated is not the same in the cases of the two works discussed, because they
were produced in two very different colonial situations. Imagery operated at a
variety oflevels in the Americas before the Conquest and to reduce it and its place
in the colonial world to the single arena of concerns about idolatry and religious
icons overlooks the dynamic capacity of native expression. The examples I have
chosen are truly New World images that reveal in one manner or another such
dynamism. They are different from the European-derived illustrations in the
works of Guaman Poma de Ayala, Sahagun, and Duran (figure 6), which are
descriptive images of the New World, because they represent a particularly native
construction of the 'newness' of the New World as a colonial entity.
The Madonna
The Mexican image comes from a thoroughly colonial context. It is one of eight
native paintings (figures 4 and 7) interspersed in a larg er manuscript, now called
the Huejotzingo Codex, of seventy-nine folios written in Spanish.' Some of the
paintings are on fig-based paper and some are on maguey-based paper,10 and
they are painted , for the most part, in the style of Aztec tribute lists , which are
composed of series of pictographs. The paintings survive among the documents
58

THE

M

AD

O N N
HORSE

A

AND

THE

in a lawsuit brought by Hernan Cortes against the members of the first Audiencia
of Mexico, Nufio de Guzman, Juan Ortiz de Matienzo, and Diego Delgadillo,
who had taken advantage of Cortes' return to Spain in 1528 to gain control of
lands and resources granted to him. Their survival as evidence in the case does
not mean, however, that the eight paintings were produced for the trial. Rather
it is more likely that they were produced during the period of Nufio de Guzman's
control of Huejotzingo as a native record of events of that period. First of all, it
is clear that they were not made all at once by the same person, because of the
variety of paper types and stylistic differences that reveal the work of at least
three different tlacuilos (scribes ). Moreover, the paintings were not originally
part of the case but were presented only after the plaintiff became aware of
their existence:
I, Garcia de Llerena , in the name of the Marques del Valle [Cortes] in the
lawsuit which I am conducting with the Licentiates Juan Ortiz de Matienzo
and Diego Delgadillo concerning the interests of the town of Huejotzingo ,
say that it has come to my notice that in this city there are certain leading men of
the said town who have paintings of what the said town gave to the said licentiates.

I beseech Your Majesty that you command and compel them to give the paintings to the secretary because I make presentation of them; and I ask that
their statements be taken by means of the paintings and by the questionnaire
which I have presented in the case and for this I ask an extension of time if
necessary... (emphasis mine)''
The paintings represent an additional form of evidence called to be presented
by the Indians of Huejotzingo to document their (an d by extension Cort es' )
costs for the enforced aid to Nufio de Guzman and the other members of the
Audiencia during their military expeditions and building campaigns." The other
form of evidence, the written document with which the paintings were included,
was recorded by a Spanish scribe from the translated oral testimony of three
native witnesses.
The written document , of course, is more comprehensive, covering the full
extent of Cortes' charges against Nuno de Guzman and the others, and includes
the testimony of many Spanish witnesses. The paintings record only the material
costs of forced tribute, labor, and purchases which are testified to in the written
document. But unlike later colonial tribute lists, such as the Codex Mendoza or
the Pintura de los Tributos de Coyoa can , the paintings are not glossed with
Spanish or Nahuatl (Aztec) text (wit h one minor exception)." Rather , the two
sources of evidence appear together equally as Mexican and Spanish traditional
forms of documentation. In fact, during the recording of the oral statements, one
59

N AT
LATIN

.

I

VE
ARTISTS
AM ERI CA

AND

PATRONS

I

N

COLONIAL

.
(

\

0

o[ll],_

(

·-

•

0

0

il ·

1-

0

0ooooo

½.'
!

Figure 7. Painting number 6, Huejot zingo Codex.
of the three native witnesses from Huejotzingo , Estevan , asked to be shown
several of the paintings, including the painting with the image of the Madonna,
and after viewing them he related the specifics of his testimony .''
It is at this juncture in Estevan 's testimony, however, that the context in
which the Mex ican painrjngs functioned shifts from a native one, in which they
are used as mnemonic devices, to a European one, in which the truth of their
contents becomes more importa nt . After having been commented on by Estevan ,
the paintings are further identified by the scribe, who interjects the authority of
his presence into the testimony by writing that 'all that has been stated is painted
on the said paper on which is depicted the said image of Our Lady and which is
60

THE

M

AD

O N

N

A

AND

T H

E

HORSE

signed with my sign.' 15 The paintings, all of which bear the scribe's rubric, thus
enter into the written testimony as material evidence establishing for any future
European audience , as the case moves to higher courts, that what has been said is
true.16 Moreover , while the paintings are not glossed, the scribe's rubric, whichis an
ambiguous sign standing between the universality of the written word and
the unique sign of an individual , signifies here the cross-ove r from the written
testimony to the paintings. " That is, the scribe's rubric both appears and is stated
in the testimony to appear on the back of each painting, to assure that this is the
painting referred to. The paintings are , as it were, ' tagged and entered .' One sees
here how neatly the ' content ' of the native signs is transformed into the signs of
the written word. They have become evidence of an act , illustra- tions of a text.
Bu t, whereas the nature of the image as sign changes through a contextual
shift, this is not a shift controlled by the native artists. It takes place through the
institution of •he Spanish legal system and all eight paintings are equally subject
to the power of this discursive transformation. Our concern here is the original
native context of the images, however. From this perspective, one finds that the
capturing of what is 'new' within the discursive strategy of ancient Mexico takes
place in only one of the paintings , the one which includes the image of the
Virgin Mary. And I mean capture quite literally. The image of the Virgin Mary is
taken over through native representation to be something 'other' than what it is
in a European context. It is changed from a Christian devotional image to a
material object through the artistic conventions of ancient Mexico.
The style and organization of this painting, like the others, follow the basic
pre-Hispanic formal conventions of a tribute list, the most 'standardized of any
form of Aztec art' (for example, figure 8). Systematically arranged in horizontal
rows against the neutral or empty space of the page, the pictographs of figures
and objects are schematic and repetitive. They are indicated by only a black ink
outline forming the contours of most of the figures, with the exception of one
row in which the interior areas of the human figures are filled by cross hatching
or concentric squares in order to indicate the textile designs of their garments.
If the painting is divided vertically more or less in half and then only the left
side is read from left to right , top to bottom, as is the normal pattern of reading
a written text, then one sees a pictorial list of the items that the Indians were
forced to contribute arranged in horizontal rows and depicted in abstracted
glyphic and pictographic forms as in other tribute lists. Following the identifications made by Lucas, the second witness from Huejotzingo, 20 there is, first, a
tied bundle of grass with a jar at the top. 21 The grass bundle is the glyph for the
Nahuatl word centzon tli , four hundred, which with the jar above it signifies the
1

'

1

'

22

6r

NAT I VE ARTISTS AND PATRONS IN COLONIAL LATIN AM E R IC A

Figure 8. Codex Mendoza , 1541-42 , folio 20.
From Codex Mendoza , commentary by Kurt Ross (Fribourg , 1978), 37.

THE

M

AD

O N

N

A AN D

TH

E

HORSE

four hundred jars of liquidambar that were supplied. This is followed by a similar glyph and pictograph representing four hundred mantles. Sixteen hundr ed
pairs of sandals are indicated next by four bundles with feet at the top. The last
object in this row is a banner with ten outlined squares, indicating ten painted
banners that had to be contributed to Nuno de Guzman. " The next row down
shows ten bundles with arrows on top, tallying up the four thousand arrows
given to the Spaniards. Below the arrows are ten glyphs each representing twenty
loads of twenty mantles or four thousand in total.
Except for the image of the banner, these glyphs and pictographs represent
objects that were common tribute items prior to the European invasion, and,
although the colonial legality of the tribute is at issue in both the native and
Spanish documents , the manner in which these specific traditional items are
represented is not problematic. It is not problematic because, as has been pointed
out, such tribute books focused pictorially on things rather than actions. " So,
altho ugh here it is the colonial action of illegal tribute that is at stake rather than
the tribute itself, it is the objects (their pre-Hispanic glyphs) that stand as evidence. As such their visual appearance coincides with their written appearance in
the Spanish text as it was taken from oral testimony; that is, the objects and quantities are simply listed sequentially in the same order as they are ordered in the
painting. The major difference is that from the written testimony one discovers
that, according to one witness at least, most of these things were used to supply
the thousand warriors forced to accompany Nuno de Guzman to war, and thus
they were not given directly to Nuno de Guzman himself.
I have described the order of the images as if they were to be read from left to
right according to what is considered to be a standardized pattern by Europeans ."
But this is not the sequential order of this painting if one takes into account
Lucas' oral testimony, as transcribed into the docume nt. And his testimony is
crucial because it is only through him that any of the paintings can be read
through the written document in a coherent manner. The other two native witnesses ask to see the paintings while testifying , but they do not systematically
replicate the order and composition of the painting in the narrative pattern of
their oral testimony."
What is significant is that, unlike the other two native witnesses, Lucas does
not actually refer to the painting as he testifies, although he, in fact, describes it.
Lucas probably had the painting placed before him as he gave his testimony (he
mentions papers [paintings] that record the tributes and things given to Nuno de
Guzman at the beginning of his testimony), but he always couches the validity of
his testimony in terms of whether he was an eyewitness to the events. Truth here

NA T I VE A RT I ST S AND PAT RO NS IN CO LO N I A L LAT I N AM E R I CA

is established by statements of what Lucas does or does not know from personal
experience and not by his reading and recaJling of information from the paintings.
The authority of his testimony is based solely on his status as an eyewitness.
Although Lucas insists on this status, the fact that his testimony follows the
painting is important, because it indicates that he too is using it as a mnemonic
device to order his recollections. " He does not ask that the paintings be entered
into the legal dialogue, as Estevan does, because they are not the standard of
proof by which he corroborates his narrative. In Lucas ' testimony the paintings
remain solely within the context of a pre-Hispanic use and do not enter as a
physical form of evidence. Estevan 's use of the manuscripts as evidence, in contrast, changes the nature of his testimony, because he enters into a dialogue with
the scribe through the paintings. That is, once Estevan asks to see the paintings,
the scribe inserts himself into the record by confirming that he too has seen the
paintings and has marked them.
Such a dialogue does not appear in Lucas ' testimony, and , if one reads the
narrative as a linear sequence and compares it to the spatial arrangement of the
objects in the painting, they do not match. Whereas one reads the text from left
to right, one must read the images from right to left with a dramatic break in
the center where the image of the Virgin Mary appears. Thus while the facts in
each form of document (the Spanish written and Mexican pictorial forms) are
almost identical , the order in which they are laid out seems to be disjunctive.
The written testimony is a transcription of an oral narrative, and its temporal
order, of course, acquires European spatial characteristics through transcription.
But preserved for us here is a reading of the images that belongs to the oral
traditions of before the Conquest. That is, the Mexican glyphic representations
in this document are not reordered according to Western ideas, even though
the oral narrative has such an order in the Spanish translation. " Lucas ' oral
testimony not only reveals the order of his viewing of the images but indicates
what that viewing meant in a native context, and, most important for this essay,
the significance of the image of the Madonna.
Lucas begins to read the painting from the right where there are twenty-one
solid gold disks arranged in four vertical rows of five with the odd disk at the
bottom. The disks represent the twenty-o ne measurements of gold that were
spent to equip Don Tome, Senor de Huejotzingo, with a horse so that he could
accompany Nufio de Guzman to war. A depic tio n of the hors e itself would
certainly have been ' new,' but the horse is here represente d by the glyphs that
stand for its value, staying within the conventions of Mexican notation. It is the
section to which Lucas refers next in his testimony where one sees the striking
image of the Virgin and Child implanted in the composition.

THE M A DO N NA AN D T H E H O RS E

Handled unlike any of the other images, the three-quarter length frontal view
of the haloed Virgin holding the Christ child is startling (perhaps 'wondrous' in
Greenblatt 's sense of the term) and out of scale. The figures have the same hard
dark outline as the others but the interiors are filled with color with various tonal
gradations to indicate the volume and folds of their garments, according to
European conventions. They are set against a dark blue background which is
surrounded by a frame of three bands of color. The frame, the solid-colored background, and the frontality of the figures isolate this image from the rest.These and
its distinctive formal qualities are based on the uniqueness of the object that this
image is meant to represent. Glyph, cipher , and stereotype cannot be used here;
this becomes a mimetic representation of a unique image. But the distinctive nature
of the representation, its not being signified by a general cipher, does not mean
that its signification stands apart from the rest of the painting. The ontological
character of the image of the Virgin Mary is transformed by its placement in the
context of the tribute list with the figures above and below it.
Only here in relation to the Virgin ' s image are the order and pattern of
discrete objects on the page broken. Lucas first mentions the two rows of human
figures below. And in this ordering, one finds that the pictorial narrative focuses
on the production of the referent, which is actually a standard with the depiction
of the Virgin Mary. The images in this part of the document are not simply
isolated ciphers tallying tribute; they also relate to each other in a temporal
sequence of actions. As Lucas notes, Nufio de Guzman requested an image of the
Virgin Mary made of gold and feathers to take to war with him, and because the
villagers did not have either the gold or the feathers to make the image, they
were forced to sell twenty slaves to Indian merchants. These are the twenty
figures shown, eight men in one row and twelve women in the other.
At this point, one is still in the realm of the expenditures to which all the
other pictographs refer. Except for gender differentiation and clothing , the
slaves are stereotypical and have no identity other than as ciphers for objects of
commerce. But unlike the other objects, such as the twen ty-one disks of gold,
their purchase value and purpose are not left to an oral narrative (which in this
instance takes on a written form in the legal testimony ). Rather the value of the
slaves is represented by what they purchased. Their exchange value is represented above the image of the Virgin by nine glyphs of feather bundles each of
which contained twenty feathers, and to the right of the Virgin by three disks
representing the amount of gold requir ed.
Value here is still represented by Mexican glyphs and pictographs of objects,
but the action of market exchange is implied by the representation of two
different sets of objects of equal exchange value. The value is represented a third

NATIVE

ARTISTS

AND

PATRONS
AMERICA

IN

COLONIAL

LAT!

time by the image of the Madonna which is the transformation of that exchange
value into representation. Lucas narrates the transformation:
And from the said ingots of gold the said image of Santa Maria was made and
it was made as broad and as long as more than half an arm. And the said nine
bundles of plumage were of the long and rich kind which the said Indians
value very highly, and they placed the said feathers all around the said image
after the manner of an enclosure . And when it had been made after this fashion,
the said lord and leading men of the said town, and this witness with them,
brought it to the said Nufio de Guzman with said overseer who is named
Gibaja. And they gave it to the said Nufio de Guzman in the presence of this
witness, and he received it."
The image in the painting refers to the Christian religious icon that the
leaders of Huejotzingo had produced and given to Nuno de Guzman to be
carried to war by him. But in this tribute list, the Christian icon has entered into
the native semiotic system through its representation. The seemingly blank,
neutral spaces between the three related images thus allow for the telling of
their histories as they are viewed. Of course the image derives from a European
pro totype , and Mexican familiarity with the image was probably dependent
upon two sources: the image of the Virgin on the standard that Cortes himself
carried into Huejotzingo and the image that was certainly later displayed to the
community in the Franciscan monastery established there.' 0 The image of the
Virgin and Child in the context of the manuscript paint ing, however, is not
simply a matter of a copy of a sacred image from a European source. That production was the task already performed by the native artisans who made the
feather standard for Nuno de Guzman. Unlike the standard, with its votive and
protective aura , this representation of the Virgin and Child seems emptied of its
Christian meaning, demythologized as it were. Although it is represented through
an analogic image and thus appears unique, its uniq uen ess is not based on the
supernatural power of the image within a Christian context; rather it is unique
because it refers to a particular object produced in Huejotzingo. Its presentation, of course, appears awkward because what is recorded is new, outside the
standards of Mexican representation. It is a presentation of an alien image that
by its likeness and, certainly, its reproduction and its size in relation to the other
figures on the page indicates a sense of wonder and display of artist ic virtuosity.
Nevertheless, the likeness , first of all, is reproduced here to signify material
value within the mode of Mexican production and exchange.
Thus the sequential order of Lucas ' testimony regarding the Virgin and the
images related to it is the pictorial representation of a narrative of economic
66

THE MADO N N A AND T H E HORSE

transformations. This then is not a Christian image but an image of an object
that happens to be a Christian image. The Madonna has been transformed,
metamorphosed through its careful surrounding by other figures, to having the
semiotic status of a cipher, a glyph. It comes to have the value of a trope within
the Mexican system.
To accomplish this act of becoming something else meant departing from
the traditional conventions used throughout the manuscript. But how was the
native artist to make his image not be read as a religious icon when that is what
the mimetic image refers to?" He (all recorded native tlacuilos are males) does
not surround it with native images merely to frame the foreign image. As simple
framing devices, these Mexican images could be read as contingent or competing
ones, in which case the artist's tactic might be seen as accidental and perhaps as
an index of his incapacity to maintain the highly structured composition of the
rest of the images in dealing with the foreign image. But this is a highly structured
compos ition that, I believe, maintains consistent standards of representation.
The norms are broken only so that the artist can contextualize the new image
through a narrative relationship with the surrounding figures. Nor is their narrative relat ionship to be read apart from the rest of the painting (and by narrative I
mean that one image relates to the next in a temporal sequence of actions ). The
narrative linking of this particular group of images therefore also serves an
intratextual purpose. The objectifying of the representation of a sacred Catholic
icon through a narrative relationship between Mexican and European images
creates a relationship of equivalence among all images in the painting.
What is therefore important about this painting is not so much that it contains one of the earliest surviving Nahua illustrations of the Madonna and Child,
a fact that no author discussing these paintings fails to mention (after all, once
the Conquest took place, there had to be a first or earliest image ). What is more
important is that the image is not just a Catholic icon but is, first of all, a Mexican
image recording its production. Whatever the residents of Huejotzingo had
learned about Christian doctrine or the veneration of Christian icons is not at
issue here . Rather , the tlacuilo has improvised on the canons of Mexican representation to convey the material cost of tribute as it took new forms. Moreover ,
it is interesting to note what the Nahuatl artists chose to depict or not depict.
The horse, a new symbol of status and power, is not represented by its likeness,
but rather a manufactured object of European representation is signified by
similitude. And that similitude is not meant to reproduce the aura of the sacredness of the Christian icon as taught in the new monastery at Huejotzingo. It
instead divests the image of its Christian iconic power and, thus, opens it to
different sorts of inquiry. One should not necessarily read this representation as

NAT I VE A RT I ST S A N D PAT RO NS I N C O LO N I A L LAT I N AM E R I CA

a purposeful act of resistance , but most surely as an act of observation and the
result of dialogue about new imagery as seen through the eyes of the people of
Huejotzingo. At the very least, we see here in microcosm a contested view of
the 'spiritual conquest ' of the New World. Because native artists were retrained
to produce liturgical objects and sacred images, as Bernal Diaz de! Castillo and
others remarked , many scholars fall too quickly into seeing their productions
and the post-Conquest New World as a matter of syncretism. This affirmative
view, unfortunately , imp lies a monolithic colonial being and disavows continued differences in relation to power - the real stuff of a colonial society - and
individual negotiations.

The Horse
The horse that is absent in the Mexican painting is the subject of representation
in one of the earliest examples of change in traditional Andean image ry toward a
colonial content. It appears as an embossed design on an early colonial gold band
(figure 5) often identified as a chipana (bracelet) because of its shape and length as
well as the two holes at the end to bind it." Again there is a juxtaposition between
traditional and new forms in an indigenous object, but unlike the Mexican painting just discussed this imagery does not articulate specific colonial interactions
between native and Spanish cultures. The Peruvian image is based on a similar
transformational operation that brings the outside world of Spanish representation into the semiotic practice of native expression. However , the transformational operation serves to situate the image within an Andean metaphoric
context." The referent is not to a specific image and act , but to general social
categories and associations . This contextual difference between Mexican and
Andean examples is not just a matter of historical chance. One cannot find a
Peruvian counterpart to the image of the Madonna; for, whereas Mexican traditional forms, such as the Huejotzingo Codex, entered into the official sphere of
colonial culture as specific evidence of facts (of genealo gy, land boundaries ,
tribute, history), Andean forms did not." Thus one must look to objects that were
made to operate solely within native Andean society to see how images of 'new '
things were improvised and 'natu ralized.' For this reason , one can rarely attach
the names of specific individuals to Andean images or find written texts to (con)textualize them. There are no Lucases or Estevans whose recorded words allow us
a seemingly more approximate understanding of the images of

the

Huejotzingo

Codex. In some ways Andean cultural practices remained more hermeneutic in
the colonial period than Mexican practices. Thus, the approximations that can
be suggested for this decontextualized Andean image are different, being, of
necessity, based on broader analogies drawn from a wider context of recorded
Andean practices and colonial institutions.
68

THE MADO N NA AND THE HORSE

The bracelet was a traditional piece of jewelry, which when made of gold
was worn by elite members of Andean society." From its design and form, it
was probably produced on the coast. The precious material and the traditional
motifs on the colonial example indicate that it belonged to a curaca or at least
someone who pretended to that rank. Curacas were the traditional leaders of
Andean communities, called ayllu, which were organized by moieties called
hanan and hurin. The curacas under the Inca provided the intermediate leadership between the imperial state and the individual ethnic groups that comprised
the empire. The Spaniards recognized the value of these ethnic leaders and they
too counted on the curacas' authority in their communities to organize corvee
labor and collect taxes for the colonial state. However, the status of curacas as
'natural lords ' ( seiiores nat urales ), as termed by the Spaniards, changed in the
colonial period and they had to be legally recognized through the Spanish power
structure. Some of the traditional signs of their 'natu ral' status also changed with
this new legal recognition, and I shall suggest that the indigenous design on the
gold band was altered to represent this new colonial status. The formal change
was from a hieratic and symmetrical design that metaphorically referred to
Andean socio-political concerns, to a design that incorporated a European-style
image that pictorially illustrated aspects of the curaca's colonial role. Yet, just as
the image of the Madonna and Child was transformed, 'captured,' to operate
within a Mexican system, so too the image of the man on a horse is drawn into
the metaphorical system of Andean representation. In both cases this capturing
and transformation must be seen as an act of appropriation. But whereas the
Mexican image (re)presents a European symbol as an object of tribute, the
Peruvian image introduces a Western object into the syntigmatic chain of
Andean metaphor .
The primary compositional structure of the band is predicated on a bilaterally
symmetrical design, but the symmetry is not fully carried out. Only the left side
of the composition is as it should be. It is composed of two horizontal rows of
alternating monke ys and felines, eight animals altogether, facing toward the
center of the composition. The monkeys are in full profile and oriented vertically
while the felines are horizontal with their bodies in profile and heads turned to
the viewer. Each vertical row of animals is composed of a male and female
member of the species, as indicated by either the presence or lack of the phallus
and testicles. In the case of the monkeys this sexual difference is accentuated by
the type of fruit each one eats. The female eats what appears to be a guama, a
fruit that has a decidedly phallic shape, and the male eats a chirimoya, a fruit that
may have female sexual connotations. The figures are metaphoric images that
pertain to a pre-Hispanic symbolic code and presumably refer to the social and
political status of whoever wore the band."
69

NAT IVE ARTISTS AND PATRONS IN CO LO N I A L LAT I N A M E R IC A

These metaphors may concern the symbolic values of the moiety division of
an Andean community, or the principles of social balance that the moieties
represent. Hanan and hurin signify a number of complementary attributes
through which the moieties were classified and, at one level of social complexity
and symbolic language, they anchor the syntagmatic chain to which further
metaphors are linked ." These social values may be represented by both the
figures and their compositional disposition. Hanan is considered male and hurin
is considered female, and this dichotomy is clearly expressed by the sexual differ ences between members of the same species. At the same time , hanan and hurin
can be expressed in terms of high and low. This concept may be represented by
the difference between the two species of the animals. The felines appear to be
pumas which are native to the highlands while monkeys are native to the
jungle. That is, the felines signify hanan's attribute of high and the monkeys
signify hurin' s attribute oflow.
One can carry this line of interpretation further by looking at the relationship
between the two sets of animals. If these anima ls do express moiety values, then
there are two representations of hanan and hurin: one by members of the same
species but of the opposite sex and one by different species from different
geographical locations. Seen together they may represent a third quality ofhanan
and hurin known as yanatin. Yanatin is a concept that reco gnizes at once the
distinction between the two moieties as well as their compleme ntari ty. Yanatin
expresses the notion of one 's social identity as having a complementary mirror
image or moiety counterpart, one's needed social and sexual opposite . This
aspect of moiety identity has been studied by Tristan Platt , who cites as an
example the ritual battles that take place in Andean communities between
moieties. " These battles not only act out the distinction between the two
moieties but they are also considered to represent symbolic coitus between the
two combatants, even though such battles are fought between members of the
same sex. It is therefore possible that the complementarity aspect of yanatin is
expressed by the sexual difference between members of the same species, while
the oppositional character of yanatin is expressed by the two different species, the
felines and monkeys. "
A specific iconographic interpretation of the animals' symbolic values within
the community in which the chipana was produced is impossible to reconstruct.
The general point, however , is that one can find close parallels between the
metaphoric language of Andean social structure and the way these animals are
represented and compositionally organized.' 0 What I wish to stress is that this
part of the band's design reveals the rich metaphoric values that were embedded

70

THE

M AD

ON N A

AN

D

THE

HORSE

in traditional Andean figural image ry, which still operate to express the transitional states of social status in Andean comm unities. Moreover, although I cannot
give a precise reading for the images on the gold band, there is clear evidence
that such animal imagery could be very specific in its references . Garcilaso de la
Vega records that there was a monumental rock painting just outside Cuzco,
consisting of two contrasting images of condors. One was depicted as a bird trying
to hide itself with wings closed and head tucked beneath one of them. The bird
faced south with its back to Cuzco. The second condor was quite different . Its
head was boldly turned toward the city and its wings were open as if in flight and
ready to swoop down on its prey. According to Garcilaso, the composition commemorated the events surrounding the defeat of the Chanca, a victory paramount
in the mytho-historic rise of the Inca empire. The first condor with its head under
its wing represented the Inca king who had shamefully fled Cuzco when the
Chanca attacked the city. The second bird symbolized his son who successfully
defended the city against the enemy."
Garcilaso's description reveals that even specific 'historical' events were represented in a broad metaphorical manner. Here 'historical' personages are subordinated to a single animal figure type whose instinctual habits stand for cultural
values; that is, the image of a sleeping bird is equated with shame and the image
of flight is equated with valor.
Garcilaso's text can also be used to introduce the right side of the design on
the gold band, where such pure metaphoric language gives way to a Europeanstyle descriptive image. In describing the rock painting he says that when he left
Peru it was in good repair, but in 1592 when he asked a priest just returning
from Peru about it, he was told that it and other images like it could hardly be
seen because they were neglected. In both literal and figurative senses, traditional imagery was fading in importance and slowly being replaced by Western
forms of representation, and it is this replacement that can be seen on the right
side of the gold band.
The far right side begins by mirroring the left with a pair of felines (presumably it originally began with two monkeys which are now missing). The second
set of four animals seen on the left side, however, is replaced by an intrusive
figure that breaks the overall symmetry of the conventionalized composition. It
is recognized at first glance as a descriptive image of a man riding a horse, a
representation that identifies the piece as obviously colonial. The horse by itself
could have been unobtrusively integrated into the composition by being replicated on the other side, or at the very least the horse could have been centered
rather than confronting the left half of the image. The human figure shown on
41

71

N AT I VE A RT I ST S AND PATRONS IN CO LO N I A L LAT I N AM E R I CA

top of the horse indicates another distinction of this part of the design. That is,
the horse is not integrated into the composition as another metaphoric element
in a purely 'natural' sense because it is combined with a human figure; it is a
man riding a horse. Only the two birds, one above the rider the other below,
signal an Andean representational context.
By placing the human figure astride the horse and controlling it by the reins
held in his hand, a physical relationship is depicted and a form of narrative is
suggested based upon a temporal and spatial interaction between two pictorial
elements . Moreover, the narrative that is suggested derives from observations
only possible in the colonial world. This figure is thus a representation of contemporary experience and at the immediate signifying pictorial level does not
connote the kind of metaphoric values that the other animals do.
But why the need to image contemporary cultural experience within a traditional composition based on animal metaphors? Its inclusion disrupts the harmony
of the composition , and the two pictorial codes, one metaphoric the other
mimetic, are awkwardly juxtaposed and seem to make little sense in relation to
one another. Yet if one heeds Ricoeur's sense of metaphor in relation to imagination as the process of 'seeing' that affects the logical distance between entities or
categories being connected, then one sees here the beginning of the metaphoric
process in which the horse and rider lose reportive mimetic value, and the semantic, cultural distance is reduced between this image and the others."
The process however does not take place simply within a closed arena of
symbolic language, but within the social and political sphere of early colonial
practice. If one presumes that the gold band through its material and form still
signified the hierarchic status of the wearer, then the two types of image do relate
because they draw the connection between the ancient privileges accorded to
the curaca by his community and the new privileges granted to the curaca by the
Spaniards in order that he be able to carry out his traditional duties.
As indicated above, the metaphoric images refer in some manner to the
traditional social position of the wearer. The associations of these symbols had
been generated and internalized through generations of lived experience. The
metaphoric images were a shared cultural language comprised of multifold
associations that rendered 'natural' and autochthonous the overt socio-political
relations of the comm unity." The 'gramm ar' that functioned at a deep level and
44

gave specificity to these associations was the structure of

that

society.

These

meanings, now lost to us, were embedded within the society as a whole and were
' intuitively' understood. In other words, these metaphors were successful and
unquestioned because they operated within an ideology that was completely
coherent and fully accepted as true.4'
72

THE

MADO N N A A N D T H E H O RS E

The figure of the man on a horse represents aspects of the Andean socius for
which there was no autochthonous image , no shared symbol, no cluster of
associations . It refers, of course, to an aspect of European power for which there
was no traditional precedent. But power as represented by a mounted horse was
not just the force of military advantage but also the privilege within colonial
society of access to a horse. And it is as much the power of privilege as the
power of force that marks the presence of the horse on the gold band."
As discussed, the curaca in the colonial period took a mediating position
between his traditional culture and Spanish society. For this role he was accorded
a number of new privileges by law so that he could fulfill his duties. In particular,
a curaca was allowed to own a horse with saddle and reins for his government
functions." This privilege compensated for the fact that a curaca could no longer
be carried about on a litter." That is, a traditional right was supplanted by a new
one, granted in order for the curaca to carry out his new duties. The horse, bridle,
and saddle were of course symbolic of the European recognition of the curaca' s
authority within his comm unity,'° but as symbols from an alien society, there
was no way for these new objects to signify within Andean society other than
through their physical presence or literal representation .
The intrusive figure on the gold band clearly illustrates the objects of the
curaca's new role. A man is astride the horse and grasps the reins in one hand
and holds either a whip or, more probably, a sword in the other (the artist 's
technical abilities perhaps restrict him here, as he shows the man's torso above
the horse and his feet directly below the belly)." The imaginary space between
culture and nature in which many Andean metaphors are generated is reduced
to domestication and mastery in a prosaic and even profane sense of power.
The only domesticated animal that served as a beast of burden in the Andes was
the llama and it is perhaps this quality of the llama that the strange 'hooves' of
the horse recall.
Thus an Andean metaphoric image could not be used here because the
relationship that was being expressed was entirely new and out of the realm of
Andean experience. Here the image seems to express the new relations only at
the level of literal illustration. " That is, the privileges that codified the curaca's
position in colonial society had to be depicted because those privileges did not
extend beyond the immediate present in a diachronic fashion nor originate from
within the structure of Andean cultu re. There was no synchronic tradition that
accompanied them because they were imposed from without. The image of the
rider therefore borrows not only the Western formula of illustration but uses
its factual content to assert the right to privilege in much the same way that
paintings were introduced in Spanish courts oflaw.
73

N ATI

VE

ARTISTS

AND

PATRONS
IN
AM
ERI CA

CO

LO

N

I A L

L AT

I N

Here, however, the image is used within an Andean context rather than a
Spanish colonial one, and to operate within Andean society such images cannot
remain external to it. They must, at the very least, be seen in relation to those
symbolic elements that constitute the representation of cultu re, if that culture is
to retain, in any sense, the power to imagine itself. That relationship is here
established through a visual language of metaphor. The simple denial in the
depiction of the feet of anatomical observation of the crucial natural sign distinguishing the equine from the camelid creates an ambivalence which cannot be
dismissed as just the unmanaged persistence of Andean artistic convention. Conventions serve to allow recognition of the familiar ; here the mixture of the familiar and the unfamiliar establishes not only ambivalence but an equivalence in
which all figures operate beyond the distinctions of natural and cultural and as
metaphors for culture, or in the case of the horse and rider as a sign for becoming
a metaphor. It is a European-style equestrian on an Andean horse. It is an ambivalence of being neither and becoming something else.
There is a dialectic, then, within the juxtaposition of two artistic systems. The
object and a part of the image harken to traditional meaning but the intrusive
Western image brings both the object and the autochthonous image into the
world of colonial reality, just as the image of the horse and rider is brought into
the signifying context of Andean expression by its inclusion on a traditional Andean
object. It stands within an Andean context and is a part of the signifying unity of
the composition, replacing while becoming an Andean metaphor. The juxtaposition articulates the contradictions of native colonial existence and tries to
reconcile the new with the old. That is, by coupling the literal image of the
curaca' s colonial position with traditional expressions of his 'natu ral' position
there is an attempt to draw an equivalence between the two." Such an equivalence conforms to a new colonial identity not only for the curaca but the entire
community. Social and political relationships are not only determined by the
ayllu structure but also by the Spanish colonial power structure as well, something that is certainly associated with the image of a horse. Moreover, this new
colonial power structure invested curacas with advantages which began to create
a social and economic division between them and their communities. The image
of the man riding the horse is only the beginning of the colonial positioning of
curacas within the world of Spanish power and economy.
But as in the Huejotzingo Codex that position is not uncontested in terms

of representation, nor in reality. It does not mean simply the mimicking of
Michelangelo by native artists. The image of slaves turned into Madonnas or
horses into nabobs is, in fact , the history of colonization. That it can be expressed
on the other side of writing, non-writing, and within terms established by native
74

THE

MA DO N NA AN D THE HORSE

Americans recognizes that colonial history was not epistemologically controlled
solely by Europeans . These images do not stand outside of the colonialization of
the Americas; they, in fact , participate in it, but they are not our images of it.
Nor are they the melancholic 'vision of the vanquished. ' They stand instead as
the capacity to (re)articulate ' newn ess' as becoming something else. And if there
seems an ambivalence in these sixteent h-cent ury images as we look at them
today, it is not based on the He gelian notion of the inherent ambiguity of the
symbolic image." It is because there was, I believe, an ambivalence then about
that becoming, an ambivalence that was certa inl y justified as the New World
became a thoroughly colonial world. Ye t, the transformative act in both examples, at the very least, turns these representations of things European into either
Mexican or Andean tropes and thereby reveals the 'other ' side of colonial discourse." Such revelations, I would suggest, are not isolated but are a constant
exercise in a colonial dialectical struggle which makes a colonial project always
and everywhere incomplete."

75

NATIVE ART I STS ANO PATRONS IN COLONIAL LATIN AMERICA

Notes

Moralise e,' in Word and Im age, v , number 1
[1989], 128); but even so there is almost an

1. For example , Ludovico Bertonio d
efines the
abstract geometric forms on Inca tex-

tiles as 'Tocapu isi: vestido, o ropa de! Inca
hecha a las mil marauillas ...' in Vocabulario
de la lengua aymara [1612] (La Paz, 1984), 357.
2.

Historia verdadera de la conquista de la

Nueva Espana, I! (1984), 455-62.
3. Jennefer Spreitzer,' Idolat ryand the Aztec
Artisan in the Florentine Codex,' unpub lished manuscript ; and Jeanette F. Petrson,' T he
Florentine Codex Imagery and the
Colonial Tlacuilo,' in TheWork of Bernardino
de Sahagun, Pioneer Ethnographer of Sixteenthcentury Mexico, edited by j.Jorge Klor de
Alva, H.B. Nicholson, and Eloise Quinones
Keber (Austin , 1988 ), 273-93.
4. The importance of writing and books in
the conquest and settlem ent of the New
World has been pointed out by many modern scholars; see most recently Walter

Mignolo, ' On Colonization of Amerindian
Language and Memories: Renaissance Theories of Writing and the Discontinuity of the
Class ical Tradition,' Comparative Studies in
Society and History, XXXIV, number 2 (1992),
301-30. However , the significance of books
and writinggoes beyond their use to in scribe

obsessive need to represent books and writin g in the New World in both Spanish and
native imagery. For examp le, one finds it in
Spanish generated imagery such as the
murals at Acto pan , the sculptures on the
facade of Yur iriapunda ro , Mexico , or La
Merced in Aya cucho , Peru, and in illustrations such as the frontispiece to Diego
Va lades ' Rhet6rica christiana . 1t is equally
important in native generated imagery such
as that found in Guaman Poma de Ayala ' s
Nueva cor6nica i buen gobierno or the Cadice de
Cuetlaxcohuapan . Finally, it often has been
noted to what extent early colo nial church
decoration is taken from book designs, and
this relation is more than just a matter of tracing sour ces. Books and buildings are metaphorically intertwined. How else are we to
understand the increasingly architectonic
structures of the frontispiece of books that
give the appearance of the portals of the
great sixtee nt h- and seventeenth-century
churches?This appearance betrays the seemingl y ambiguous use of the Spanish term
'portada ' to designate facades of structures,
ephemeral triumphal entrance arches into
citi es, and frontispieces.They are all liminal

the colonial process. The act and object
become iconic sign ifiers of the presence of
the Europeans in America. It is, after all, the

passages and openings into European worlds
and words.SeeTom Cummins ,' Boo ks, Bibles
and "Portadas," ' unpublished manuscript.

bringing of a book, the Bible, through which
legal cause was given to the Conquest. As a

5. These are the categories ofTzvetan
Todorov's extremely influential book The

sym bol of power and knowledge, the Bible

Conquest of America (New Yo rk,1984). Stephen

had long been used as an image since the

Greenblatt's Marvelous Possessions (Chicago,
1991) also casts knowledge of the New World
in terms of written texts which tell of only

Middle Ages (see Michael Camille, 'Visual
Signs of the Sacred Page: Books in the Bible

THE MADONNA AND THE H O RS E

European views. Both are literary scholars,
so it is no surprise that writing is not only

9. The document is now in the Manuscript
Division of the Library of Congress and is

privileged but also the only sourceof representation. Thus Greenblatt can reduce the
initial European response to the New World
to an essentialist argument through what he

currently known as ' The Harkness 1531
Huejotzingo Codex.' The entire manuscript
has been transcribed and translated by
j. Benedict Warren. See Howard Cline ,'The

reads: ' Wonder is, I shall argue, the central
figure in the initial European response to the
New World, the decisive emotional and intel-

Har kness 1531 Huejotzingo Code x,' in The
Harkness Collection in the Library of Congress
(Washington, 1974 ), 9-2m lt has previously
been called Document I of the Harkness
Collection (see Howard Cline, 'Ozto ticpa c

lectual experience in the presence of radical
difference .. .'(page14).
6. Rolena Adorno's Guaman Poma: Writing

Lands Map ofTexcoco 1540,' The Quarterly
journal of the Library of Congress, XXIII,
numb er 2 [1966] , 78-79) and also the Codex

as Resistance in Colonial Peru (Austin, 1986 ) is
one of an increasing number of studies by
literary historians that address the capability of a native to use European literary tech-

Mont eleone by the first modern author to
take note of the manuscript (see Alfonso
Toro, 'C6dice de! Archivo de las Duques de

niques to form the image of the European as
the 'other .' Moreover, she recognizes the
equal importance of the non- litera ry, visual
elements to such a task . See also her discussion through native textsofTodorov and
de Certeau's work ('Arms, Letters and the
Native Historian in Early Colonial Mexico,'in
1492-1992, Rel Discovering Colonial Writing,
edited by Rene Jara and Nicholas Spadaccini
[Minneapolis, 1989], 2m -24).

Mont e leo ne ,' Anales del Museo Nacional de
Arqueologia, VI [1925 ], 60 -66).
ro. For a description and material analysis
of the paintings , see Silvia Rodgers Albro
and Thomas C.Albro, ' The Examination
and Conservation Treatment of the Library
of Congress Harkness 1531 Huejotzingo
Codex,'JAIC, XXIX (1990 ), 97-n5.

n. All translations are by Benedict Warren

7. George Ku bler, ' On the Colonial Extinc-

(in Cline' H uejotz ingo Codex,' 85).

tion of the Motifs of Pre-Columbian Art,' in

12. There is internal evidence in the document that further suggests that the paintings
were not part of the original construction of
the case against Nuno de Guzman and the
others.The series of questions, or questionnaire , are referred to by Garcia de Llerena as
already presented to the court. The order
and composition of this questionnaire
(pages 91-95) does not conform to the order
or specific information in the paintings. If
the ques tio nna ire had been prepared in

Essays in Pre-Columbian Art and Archaeology,
edited by Samuel K. Lothrop and others
(Camb ridge, 1964 ), 14-34.
8. By focusing on this moment of crisis, or
even rupture in Foucault's sense of the term,
I will suggest thatTodorov 's thesis about the
inability ofN ative Americans to imp rovise,
and hence the inevitability of the European
conquest , is not true, at least not in terms of
visual communication.

77

ATIVE
ART I
LATIN A
M
ERI CA

S

T

S

AND

PAT

RO

N

S

IN

COLONIAL

relation to the paintings, one would expect a
much closer correlation between the two.
The native notion of recording the costs of
aid ing the Europeans for some future

17. The equ ivalence between iconic image
and written wo rd is alrea dy understood
within such legal documents in which one
swears to the truth both in word and image,

reference is not altogether unusual in the
early colonial period . For example, the
Wanka of Peru also kept record by quipu

'so br e todo lo cual pido me sea echo entero
conplimento de justicia e juro a Dios ya esta

(knott ed colored string) of the amount of
material aid they had supplied to their
European allies. The information from the
quipu was admitted as evide nce in a later
litigation against the Spaniards in order to
recover the Wanka 's expenditures. See John
Mur ra, ' Las etnocategorias de un khipu
estatal,' in Fomiaciones economicas y politicas en
el mundo andino (Lima, 1975).
13. The sixth painting has the words 'honze
casas princ ipales' with the scribe's rubric
written along the numerical and house
glyphs in the upper right hand comer (pages
64-65 ). They are all numbered on the back
and include the scribe's rubric; Paintings1, 7,
and 8 have annotations also on the back.
14. The number of native witnesses called to
testify is not an arbitrary numb er. Three

native witnesses were the minimum number
needed to verify a statement.
15. Cline,' Huejotzingo Codex,' n9.
16. European style portraits and paintings of

historical events were also entered as evidence incourt testimony in New Worldcases,
and their visual content was attested to by
the scribe. See Tom Cummins, ' We are the
Other: Peruvian Portraits of Colo nial Karaka k u n a ,' in Transatlantic Encounters: Europeans and Andean s in the Sixteenth Centu ry,
edited by Kenneth J. Andrien and Rolena
Adorno (Los Angeles and Berke ley,1991), 217.

+ que lo pido no es de malicia,' (Cline, ' Hue jotzingo Codex,' 74, emphasis mine).
18. Esther Pasztory, Aztec Art (New Yor k,
1983), 206.
19. Paszto ry, AztecArt , 206.
20. Lucas gives the most thorough identi-

fication of the objects and quantities in the
painting (Cli ne , ' Huejotz ingo Code x,' 10515). Where other witnesses substa nt ially

differ from his identifications will be noted
in footnotes.
21. Cline, ' Huejotz ingo Codex,' 105.
22. For example , in the 1570 dictionary of

Molina the term centzontli is translated as
'quatrocien tos, o una mata de ortaliza, o de
yer ba' (Fray Alonso de Molina, Vocabulario en
lengua castellana y mexicana y mexicana y
castellana [M exico , 1977], 18).
23. The image of the banner and the ten
squares is i nt erpreted differently in the
testimony of Estevan: 'and in order to make
a banner which he says is painted in the said
painting , so that the Indians could carry it,
they spent ten loads of little blankets for
co ve ring , with twenty blankets in each
loa d. He then corrected himself to say that
Don Tome carried said banne r' (Cline ,
' Huejotzingo Codex,' u 9).
24. Paszrory, Aztec Art, 20 6 .
25. The standardized left to right reading of
tribute lists is used on another later tr ibute /census list from Huejotzingo (see

THE MAD O N N A A N D THE HORSE

Hanns Prem ,Matricula de Huexotzinco [Graz,

205).The distinct order of this very early doc-

1974]). The left to right order is also the one

ument therefore may represent the original
organization of such tribute lists, at least
from the area ofHuejotzingo.
29. Cli ne , 'Huejotzingo Codex,' 100.

used by modem scholars to orient a viewer
in relation to a listing of the objects depicted
in the painting (see Cline , ' Huejo tz ingo
Codex,' 62; and Toro ,'Cod.ice de! Archivo').
26. For example , Estevan first mentions the
gold spent to buy a horse for the cacique and
then asks for the painting in which the gold
as well as the image of the Virgin Mary
appear (also already described) to be shown:
'And he asked to be shown another paper
where they have them [gold ingots] painted, on which is said image and painting of

30. The image of the Virgin and Child and
the symbol of the cross were the first two
Christian images that were distributed to
native communities by Cortes. See Richard
Trexler, 'Aztec Priests for Christian Altars,'
in Church and Community, 1200-1600 (Rome ,

1987 ), 469-92.

31. Of course the artist cannot control the
reading of the image especially once it
appears in a bicultural context, such as the

SaintMarywhich hehas mentioned ...'(Cline,
' Huejotz ingo Codex,' n 9). It is the voice of
the scribe which here described the appearance of the painting as a piece of evidence.

lawsuit. Hence , when the scribe records
what he sees, he recontextualizes it through
his written commentary: 'All this that has
been said is painted on the said paper on

This type of collaboration between the
scribe and a particular witness occurs whenever the paintings are produced.

which is depicted the said image of Our Lady'
(Cline, ' H uejotzingo Codex ,' 119, emphasis
mine). Here it becomes again an image of the

27. Often the order of components of the
question asked structures the order of the
information in the answer. This is not the
case here (question 4 and response).
28. It may be that the presumed standard-

Virgin Mary rather than being an image of an
image of the Virgin Mary.
32.

The identification of the piece as a

bracelet was made by Pal Keleman ( Art of the

ized organization of reading from left to right
and top to bottom of tribute lists is actually
the consequence of an early adaptation to
European patterns. By conforming the pictorial organization of colonially produced
tribute lists to the structure ofa written page,
a standardization was achieved that permitted easy comparison between the two forms.
This would expla in, in part, why these types
of documents are, of all Mexican picto graphic material, 'most frequently described
as "writing systems " ' (Pasztory , Aztec Art,

Americas, Ancient and Hispanic [New Yo rk,
1969], 175).
33. Other than the basic quality of metaphor
as the substitution of one image (verbal or
visual) for another to express a concept, I do
not assign a fixed meaning to metaphor as I
use it in this discussion. Fixing meaning to
such an elusive concept is an academic and
philosophical exercise. Metaphors, I would
suggest, can be situated without specificity
to ego. It is the engagement in play that is
important because it permits alternatives

79

N ATI
A
M
E
R
IC
A

V

E

ARTISTS

ANO

PATRONS

IN

COLONIAL

LAT

I

N

39. See , for example , Gary Urton, 'Animal

even as it employs categories of definitions.
l thus have employed a variety of different

Metaphors and the Life Cycle in an Andean

modem comments about metaphor in relation co my discussion of chis piece, not in
order co build up a single or unitary definitionofmetaphor ,but ratherto suggest the var-

Comm unity,' in Animal M ythsand Me taphors ,
2 51 -84.
40. This kind of social identity of difference
and complementarity certainly conditioned
ritual behavior imposed by sacred architecture on the central coast well into the colo-

ious possibilities of the playamongimages.
34. For differences in Span ish reactions to
and uses of native imagery in the viceroy-

nial period. Felipe de Medina described in
1650 a temple complex in use near Huacho
that must have been ve ry similar co the
ancient pre-Hispanic temple at Chavin de

alties of New Spain and Peru, see Tom
Cummins, ' Repres entation in the Sixteenth
Century:The Colonial Image of the Inca,'
Dispositio, forthcoming.

Huancar. le was a U-shaped structure with
underground chambers that led co a central
image 'de piedra muy extraordina ria

35. The material of an object in Andean culture waspare ofche symbo li c con tent expressing rank and status; see Guaman Poma

... traido de muy le jos.'The figure was over
nine feet tall and carved in relief with deep
spirallin g channels that emanated from the
eyes and through which ran blood and chicha sacrificed to the image. All those who
sacrificed came together in this central
chamber ; however , their paths to it were
conditioned by opposing ethnic and sexual
id entiti es.There were four separate stonelin ed underground passages, two for people from the coast and two for people from
th e sierra . Men from the sierra entered
through one and women from the sierra
entered through the other; the same divi sion was made for people from the coast.
Thus, one finds the same divisions imaged
through animal metaphors on the chipana
as operating in ritual performance , where
opposites became reunited in a single central space. See Felipe de Medina ,' Rela cion
del .. .Visitador Genera l de las idolatrias del
Arzobispado de Lima ... ,' in Colecci6n delibros

de Ayala , Nueva cor6nica i buengobierno,
edited byJohn Murra and Rolena Adorno
(Me xico, 1980 ), 464-65.
36. For the ln caic use of animal imagery in
relation to symmetry as a political and social
metaphor, see Tom Zuidema,' The Lion in
the City,' in Animal Myths and Metaphors in
South America, edited by Gary Urton (Sale
Lake City, 1985), 89-95.
37. The importance of the social categories
ofhanan and hurin asgeneratingsomeof the

centra l metaphors of Andean expression
was first introduced in a significant manner
by Tom Zuidema ( T heCeque System of Cuzco
[Leiden , 1967] ) and has been used as an explanatory model for a number of studies of
Andean representation. See , for example,
Rolena Adorno's discussion of Guaman
Poma's drawings in Guaman Poma: Writing
and Resistance.

38. My discussion of yanatin in general is derived from his work ( Espejos y maiz:Temas de
la estructura simb6lica andina [La Paz,1976]).

y documentos referentes a la historia del Peru
(1650 ], second series, III (Lima , 1920 ), 89-91.

80

THE MAD O N N A AND THE HORSE

Thecompositional use of male and female
as symmetrical elements with both comple-

representation, " which specifically consists
in the belief that the social phenomena in

mentary and antithetical metaphoric associations has an extremely long history in
Andean representation beginning in the

question are nonsocial or "natu ral" in origin

Early Horizon . See Patricia Lyon , ' Female
SupematuralsinAncientPeru,' NawpaPacha,

phors,50 ).

or essence'('Animal Symbolism ,Totemism ,
a nd Structure,' in Animal Myths and Meta-

41. See, Urton , 'Animal Metaphors.'

For a detailed study of the manner in
which symbols work within a native society at both a surface and deep level and the

42. Los comentarios reales de la s Incas

(Barce lona, 1968 ), Book IV, chapter 23.
43. Billie Jean Isbell, 'The Metaphorical

process by which the two levels are fully relat ed and understood, see Gerardo Reichel Dolmatoff, Amazonian Cosmos (Chicago and

Process: "From Culture to Nature and Back

London , 1974 ). Nonethe less , the representation of the social order as a ' natural' order
through the use of metaphoric images is the

46.

XVI (1978 ), 98-103.

Aga in,"' in Animal Myths and Metaphors, 301;
Paul Ricoeur, 'The Metaphorical Process as

obfuscation ofreality in order that a society
mayoperate to overcome contradictionsthat
might otherwise destabilize social relationships.SeeT urner,'Animal Symbolism ,' 49-54.

Cognition , Imagination , and Feeling,' in On
Metaphor, edited by Sheldon Sacks (Chicago

and London , 1979 ).
44. That chipanas or bracelets continued to

47. This mimicking is what Homi Bhabha
calls the 'me tonymy of presence' in his discussion of early modern Western colonial
disco urs e.Although discussing British textual examples of the nineteenth century,!
would suggest that his process operates in
many colo nialsituationsin cluding theexam ple discussed here . These instances of metonymy are the nonrepressive productions
ofthe cultureofenunciation through astrategic confusion of the mean ing. For each of
these instances of 'a difference that is almost
the same but not quite' inadvertently creates
a crisis for the cultural priority given to the
metaphoric, as the process of repression and
substitution which negotiates the difference
between paradigmatic systems and classifications. In mimicry, the representation of
identity and meaning is rearticulated along
the axis of meto nymy . As Lacan reminds us,

have significance for curacas is evidenced by
their being listed by their Quechua (Inca )
name in wills. For example, in the 'testa mento de Don Pedro Arapa,cacique [curaca J
principal de la parcialidad de los Indios Collas deste pueblo de Pocona' Uurisdiction of
M izq ue ), are listed 'dos chipanas de oro
grandes,' 'dos chipanasde oro chiquitos,' and
'una chipana de plata ' (Arch ivo Municipal
de Cochabamba / Ramo Mizque, volume
1561-90, Expedience number 3). I would Like
to thank Lolita Gutierrez Brockington for
sharing this document .
45. Social facts are made' natur al' through
the representation of animals or other
natural phenomena as symbols. That is, as
Terrance Turner in rephrasing Mclennan 's
1870 concept offetishism writes:'the representation of social or cultural phenomena in
animal form involves an element of "mis-

8I

NATIVE ARTISTS AND PATRONS IN COLONIAL LATIN AMERICA

'mimicry is like camouflage, not a harmo-

Provincia de Quispicancha ,' fo lios 735-41

nization or repression of difference, but a
form of resemblance that differs/ defends
presence by displaying it in part metonymi-

verso; and ' Inventario de Cajamarca 15-Vll-

cally' (Homi Bhabha, 'Of Mimicry and Man:
The Ambivalence of Colonial Discourse,'
October, XXVIII [1984],130-31).

de Condebamba y Provincia de Cajabamba,'

1647,' folios 12- 15, cited in Waldemar Espinosa Soriano, 'Los Sefiorios etnicos del Valle
Anales cientificos,III (1974), 133.
51. The right to carry arms wasanother privilege accorded only to curacas ( Espinosa,

48. This law was repeatedly stipulated in
varying forms in the ordenanzas of Francisco de Toledo. See Francisco de Toledo disposiciones gubernativas para el Virreinato del
Peru 1576-1580, II (Seville, 1989), 242-45.

49. Gregorio Gonzales de Cuenca, 'Ordenanr;:as delos Yndios,1566,' HistoriayCultura,
IX (1975),145.The image of a figure carried on
a Litter becomes a sixteenth andseventeenthcenrury motif to represent the despotic power of the non-European ruler (both Asian and
Am erican ) who greets and then often is
justly overthrown by Europeans. The earliestimage ofanAndean offered to Europeans
was Atahualpa carried on a litter in the act
of receiving the Bible from Fray Vicente
Valverde (in Cristobal de Mena's La conquista
del Peru [Seville, 1534]). The image immediately precedes Atahualpa's act of refusal
of the book which gives the Spaniards just
cause to pull him from the litter and to seize
the empire . Ultimately, as a trope, the image
of the litter becomes a site of critique, as it is
used by Montaigne in hisessay, 'Descoches.'
SeeTom Conley, ' Montaigne and the Indies:
Cartographies of the New World in the Essais , 1580-88 ,' in 1492-1992: Rel Discovering
Colonial

Writing.

50. For examples of the prestige of passing
on such objects, see Protocolo 2/2 28, Escribano Gregorio VasquezSe rrano,'Tes timonio
de Don Joseph Aocatinco de Quiq ijana ,

'Sefiorios etnicos,' 131) and was often stipulated in the cedulas granting a coat-of-arms
to a curaca, such as the one granted in 1598 to
Jua nAyaviri:'Que sedecedula para que tenga
armas de pasados y pueda traer espada y
daga y lo mismo pueden hacer sus descendientes .. .' ('Memorial de Charcas,' cited in
Silvia Arze and Ximena Medinaceli, Imagenes
y presagios, el escudo de los Ayaviri, Mallkm de
Charcas [La Paz, 1991], 12). In the' testamento
de don Pedro Arapa , cacique principal de la
parcialidad de los Indios Collas ' (cited in
note 44) not only are chipanas listed but also
'dosespadas con guamaci6n de atauxia.'
52. Thereisadirect correlation herebetween
indigenous imagery and mythology. The
same native thought that employs myth
and creates image is based on the concrete
and not abstraction (Claude Levi-Strauss ,
The SavageMind [Chicago and London,1966],
2 2-30). Metaphor is used here to express a
fixed relationship. The metaphoric symbol
may change from one myth to another , but
the relationships that are expressed in the
myth do not. In other words, there is an
underlying structur e in mythology , like
grammar , so that symbols in myths have no
intrinsic significance but are dependent
upon their context (Levi-Strauss , The Raw
and the Cooked [ New York, 1975], 56). The
Spanish Conquest brought an entirely new

82

THE MADONNA AND THE HORSE

setof relations for which there was no form

also the critique byJohannes Fab ian , Time

for expression other than concrete representation at a surface level.

and the Other (New York, 198 3), 123-31.

53. The same difficulties arise in native literature in which Andean and European traditions are brought together. See Frank
Salomo n, ' Chronicles of the Impossible,' in
From Oral toWritten Expression: NativeAndean
Chronicles of the Early Colonial Period (Syracuse, 1982 ), 9-12.
54. Georg Hegel, Lectures on Aesthetics. See

55. Although speaking about langu age , I
think Ricoeur's statement that ' by providing a kind offigurability to the message, the
tropes make discourse appear ' can apply to
the images discussed here ('The Me taphorical Process as Cognition, Ima gination , and
Feeling ,' in On Metaphor, 142 ) .
56. I would like to thank Simon O'Meara for
a close reading of the draft of this paper.