OLMEC COLOSSAL HEADS AS RECARVED THRONES:
'MUTILATION', REVOLUTION AND RECARVING.
James B. Porter
Archaeological Research Facility Department of Anthropology University of California Berkeley June 11, 1989
.
When Jose Maria Melgar y Serrano stood amazed before the first of the famous Olmec colossal heads more than 120 years ago, he could not have known the bitter archaeological debates his discovery would fuel and continues to feed to the present day. Various authors have subsequently noted that sculptures such as Melgar's colossal head are a primary defining feature of the archaeological Olmec culture (de la Fuente 1973, 1975; Graham et alii 1979). This situation subjects to often acrimonious debate such fundamentals as the origins of the style, the historical relationships of Olmec sculpture to other Mesoamerican sculpture traditions and the ethnic identity, linguistic stock and social organization of the producers of Olmec sculpture. Even the name Olmec - Aztec for 'rubber people' - is borrowed from an unrelated people who occupied the Gulf region during the sixteenth century. All that is certain about the archaeological 'Olmec' is that they occupied parts of southern Mexico and Guatemala during the last few millennia BC. Despite these problems, there are a number of different facets of Olmec sculpture which may be profitably examined. One of these appears in a group of Olmec colossal heads and 'altars' which bear evidence of a set of modifications, some of which also appear on other Olmec sculpture types (Grove 1981). Though the purpose of Olmec sculptural modification remains unknown, direct examination of specific monuments suggests that such modifications may shed light on a complex pattern of Olmec sculpture reuse.
To identify the original monument type represented by 'arcs' above the ears of these monuments, the heads must be turned 90 degrees onto their backs. When this is done, the carved 'arcs' are seen to be the upper edge of a niche framing the shoulders and armpit creases of an effaced human figure. The upper edge of the niche is interrupted by the remnants of the head of the figure which once projected in high relief or half round from the niche. The distance from the niche figure's shoulders to the colossal head's flat back is sufficient for the body of a complete niche figure. Large flake scars on the right side of San Lorenzo Head 7 show where the niche figure's head, forearms, legs and lower body were broken away prior to recarving. The niche figure's upper torso and the top of the niche survived recarving because they were deeply carved in sunken relief, which is harder to efface than raised relief or sculpture in the round.
Niche figures such as these, while common in Olmec art, are most prominent as the central feature of the front of subcubical Olmec 'altars'. The sunken relief shoulders and upper torso of these niche figures are usually framed by an arched niche, while the head projects from the niche in high relief or full round carving. The presence of effaced niche figures centered on one of the long sides of San Lorenzo Heads 1, 2 and 7 reveals these sculptures to be 'altars' which were later recarved as colossal heads, whose previously puzzling flat backs are now clearly seen to have been the flat underside of the former 'altars'. The sculptural stratigraphy represented by sequent phases of carving on these three colossal heads reveals a pattern suggesting the fascinating probability that many colossal heads are also recarved 'altars'. Only Tres Zapotes Head 2, Cobata Head 1, San Lorenzo Head 8 and Abaj Takalik Head 1 (Monument 23), do not show the rounded subcubical shapes which are consistent with recarving from 'altars'.
Virtually every student of Olmec art has noted that there are two forms of colossal heads: 'round or spherical' and 'elongated'. There are also two basic forms of 'altars': 'square or cubical' and 'elongated rectangular'. The 'round or spherical' head derives naturally from the 'square or cubical' 'altar', and the 'elongated' head from the 'elongated rectangular' 'altar'. The significance of this observation becomes clear when it is noted that La Venta 'altars' are cubical, apart from the rectangular profile of La Venta Altar 4, as are La Venta heads 1 and 4, while San Lorenzo 'altars' are elongated - as are the San Lorenzo
colossal heads. The shape of Tres Zapotes region heads was attributed by Howell Williams to locally occurring spherical boulders. Tres Zapotes Head 2 and Cobata Head 1 do not have flat backs and they may not be recarved
'altars'. However, both these heads have flat bottoms, suggesting the possibility that the heads could have been carved from 'altars' which were not turned on their axes. Also, it should be noted that Tres
Zapotes Head 1 has the characteristic flat back which suggests a recarved 'altar'.
The evidence that the two small cubical 'altars' from Laguna de los Cerros were being recarved suggests an interesting regional variation on the reuse pattern. Colossal heads as such have not been found at Laguna de los Cerros, though two large cubical grotesque heads are known from the site. It is possible that these two grotesques are the local equivalent of colossal heads at other Olmec sites and were recarved from small cubical 'altars' similar to the two found at the site. A similar situation may also exist at Cerro de las Mesas (Monument 2) and Tres Zapotes (Monument 25) where, though no 'altars' are known, flat-backed abstracted and mask-like colossal heads have been found. Indeed, the Tres Zapotes colossal head is clearly unfinished and bears traces of a niche on its back (Porter 1989: 135-137). If the grotesques at Laguna de los Cerros and the Cerro de las Mesas and Tres Zapotes heads were also recarved 'altars', then the conception of colossal heads as large portrait-like human heads does not strictly correspond to the actual Olmec functional monument type. Therefore a more accurate designation of these sculptures might focus on recarving from 'altars' rather than on colossal size or portrait-like naturalism.
Patterns in the appearance of standard sculpture modifications also illuminate the recarving of 'altars' into colossal heads. Coffering appears on 'altars' (La Venta 'Stela 1', La Venta Altar 4 and San Lorenzo Monument 14) and on San Lorenzo Head 2, which was once an 'altar', suggesting that coffering was confined to 'altars'. The half-effaced coffers on the curve of the back of San Lorenzo Head 2, in what was the original 'altar''s bottom and front edges, further suggest that coffering was added after obsolescence and before 'altars' were recarved as colossal heads. Also, a lowered surface on the back of this head 'may have been purposefully carved, since in Monument 14 from San Lorenzo [coffers] were carved in a new surface of the head created after carefully chipping away earlier decoration' (Clewlow et alii 1967: 79). However, the presence of a small coffer on the otherwise unmodified front of La Venta Altar 4 suggests that coffering may have preceded the chipping away of earlier decoration. Also, these modifications are clearly not 'mutilation' in the sense of defacement, since significant elements of relief on both the La Venta and San Lorenzo 'altars' survive intact. Cupping is the most common modification to the colossal heads and may have been the final stage in surface modification. Grooving appears on every kind of Olmec sculpture and is therefore difficult to place within any consistent carving sequence.
At Abaj Takalik and possibly at Izapa there are notable contrasts to the foregoing sequence of recarving from '`altar' to head'. Abaj Takalik Head 1 (Monument 23), as noted previously, is almost certainly not recarved from an 'altar'. It provides, instead, an example of a colossal head (with cupping on the left side) where the original head's facial features were recarved into a seated niche figure. The recarving of this head is quite clear, despite the claim in a recent publication that the ears of the colossal head are actually the 'eyes' of an 'open-jawed monster' (Parsons 1986: 10, 19). Izapa Miscellaneous Monument 2 may also represent a flat-backed colossal head with its facial features recarved into a squatting niche figure.
Including coffering and other modifications, evidence of six sequent stages of carving survives on Abaj Takalik Head 1:
Including grooving and other modifications, evidence of six sequent stages of carving survives on the rounded La Venta Head 4:
Including coffering and other modifications, evidence of six sequent stages of carving survives on the elongated San Lorenzo Head 2:
Grove provides a reasonable basis for explanations of colossal heads as recarved 'altars', in the only plausible interpretation of Olmec 'altars' yet proposed, by suggesting that 'altars' were actually thrones (Grove 1973). If his interpretation is correct it is likely that such thrones played a role in the careers of Olmec leaders commensurate with their impressive appearance as sculptures. Obsolescence following the inauguration, jubilee or other events for which the thrones were carved would have made it possible to convert these impressive monuments into another kind of commemorative sculpture such as the colossal heads. Possibly these colossal heads are portraits or effigies of the leaders for whom the thrones were carved, and they may have been converted into mortuary monuments following the leader's death. However, from the existing archaeological and sculptural record it is not possible to determine whether throne recarving occurred before or after a leader's death. The relative distributions of thrones and colossal heads may also prove significant in this context. Two thrones have been found at Laguna de los Cerros, three at San Lorenzo and as many as nine from the environs of La Venta, all with features removed to commence the process of recycling. This suggests the possibility that more than one throne may have been used at one time, or perhaps that multiple authorities were entitled to official thrones. In-group change could also have provided a motivation for recycling the thrones of previous regimes. Only at La Venta do thrones outnumber colossal heads. There is a scarcity of surviving thrones at San Lorenzo, Laguna de los Cerros and Abaj Takalik, as well as an absence of thrones in the existing sculptural corpus at Tres Zapotes. All of these factors suggest that thrones may not have had a long use life. Furthermore, comparison of the large ratio of thrones (Altars 1-8 and 'Stelae' 1 and 5) to colossal heads (1-4) from the environs of La Venta with the small ratio of thrones (Potrero Nuevo Monument 2, San Lorenzo Monuments 14, 18 and 20) to colossal heads (1-9) from the environs of San Lorenzo implies that San Lorenzo's officials often endured long enough to recycle their altars into colossal heads, while La Venta's officials did not. Finally, differences in sculptural practices at different times and sites, as well as accidents of preservation and discovery along with unusual historical events may all have skewed the extant monument sample.
Final resolution of all issues raised by the observations and proposals presented here must await further documentation of Olmec sites and sculptures. Until such thorough records are available, only preliminary hypotheses may be proposed to explain the recarving of Olmec thrones as colossal heads, and a broad perspective of alternative possibilities should be kept in mind before embracing any single and simple scenario for explaining the variety of Olmec sculptural practices. Previous examinations of Olmec sculpture have been articulated by an anthropological emphasis upon economic and technological aspects of the record, often to the exclusion of other relevant factors. This theoretical bias frequently results in a cavalier treatment of artistic issues in anthropological studies of the Olmec and other indigenous Mesoamerican civilizations. Indeed, without accepting his diffusionism, I must agree with Paul Schau's plaint that 'visual illiteracy or, at best, visual insensitivity [characterizes] the field of Olmec archaeology' (Schau 1983:337). Mesoamericanists have traditionally treated flat-backed colossal heads and broken thrones as separate problems. Flat-backed colossal heads have mostly been treated as an aesthetic or as an technologically artistic issue, while broken and 'mutilated' thrones have always been treated as a social-political issue. Stirling (1955:20) originally explained the flat-backed colossal heads as resulting from the aesthetics of placing them against some structure. However, Clewlow (et alii 1967:66-7) noted the absence of such structures in the archaeology of colossal heads and attributed the flattened backs to unspecified 'stylistic' motives. Heizer explained the flattened backs of colossal heads as an anticipation of the use of rollers and other technological aids for transporting the heavy sculptures (Heizer: personal communication). Furst and Furst (1980) subsequently argued, that the faces of colossal heads were flattened to reduce breakage in transport. Coe and Diehl (1980) carry this kind of economic determinism to an extreme, asserting that the flat 'negroid' faces of Olmec colossal heads result from a desire to save the labor costs entailed in carving ''typical' American Indian' portraits. Yet, the laborious removal of what is calculated to be more than half the mass of a subspherical natural boulder, to form first a subcubical throne and finally again a subspherical head shows clearly the ethnocentric and materialist naievete of such a judgment. Furthermore, the flattened backs of Tres Zapotes Head 1 and Abaj Takalik Head 2 suggest they were also recarved 'altars'; so recarving of altars was not confined to sites where stone was not readily available, Tres Zapotes being very near a source of workable stone while at Abaj Takalik workable stone actually occurs within the ruins. In any case, among the most important motivations in the production of monumental sculpture are the ostentatious display of wealth, artistic ability and religious/ political power. Since recycling impressive monuments in which a great investment has already been made only decreases the total inventory of monuments for display, it is reasonable to look elsewhere for explanations.
Stirling, followed by Drucker et alii, attributed broken thrones to acts of iconoclasm by violent successors of the Olmec (Stirling 1940:334; Drucker, Heizer and Squire 1959:230). Later, following Thompson's (1954) interpretation of the Classic Maya Collapse as the result of a class struggle, Heizer suggested violent revolution as an explanation for broken Olmec 'altars' (Heizer 1960:220). Coe subsequently embraced Heizer's and Thompson's evocation of Marx's social theories: he too explains broken Olmec monuments as a result of violent revolution (Coe 1967:25, 1968:63). Coe's claim that a violent revolution at San Lorenzo is evidenced by the 'mutilation' and 'burial' of sculptures in alignments, 'in one great act of destruction' (Coe and Diehl 1980: 387) has been widely accepted in texts and syntheses of Mesoamerican culture history. Indeed, some writers have used it as a point of departure for even wider ranging interpretations. However, this 'revolution hypothesis' stems from a misunderstanding of what constitutes mutilation. A careful examination of the undefaced niche figures of broken San Lorenzo thrones should have amply signalled that recycling, rather than 'mutilation', was the intent. Nor does careful examination of the final San Lorenzo report sustain the claim for the simultaneous 'burial' of the 'mutilated' sculptures (Graham 1989:240-2, 244-6). Also contraindicated is the recurring claim that Olmec heads were carved off site (Coe and Diehl 1980:297; Furst and Furst 1980:14). Finally, the recycling of Olmec sculptures not only at San Lorenzo but at La Venta, Laguna de los Cerros, Abaj Takalik and other sites as well, clearly indicates a widespread tradition of recarving through time.
Grove has criticized this 'revolution hypothesis' in the only previous study of broken Olmec monuments which perceptively examines the monuments themselves. He also departs from previous methodology by taking a social anthropologist's approach. Grove interprets broken monuments as protection from the supernatural powers of dead rulers, through an ethnographic analogy 'drawn from the belief systems of present-day Indian groups of tropical-forest South America' (Grove 1981: 67-8). His use of ethnographic analogy is an interesting departure from the purely theoretical approaches of his predecessors.
Archaeological facts useful for the interpretation of art are scarce in Mesoamerican archaeology, and one test of a significant discovery is the extent to which old ideas, and even 'facts' are rendered obsolete. The foregoing discussions of flat-backed colossal heads and broken thrones, share an overreliance on economic and technological interpretive approaches. They also rely on superficial and incomplete visual examinations of the Olmec sculptures themselves, despite the fact that some of these have been available to study for more than fifty years. Surely more fruitfull results can be achieved by critically reconsidering the sacrosanct status of established archaeological reconstructions and basing future explanations on the direct and detailed observation of materials actually under examination. The present study employs such a program in identifying the flat backs of Olmec colossal heads as the simple survival of a fundamental feature of the original sculpted form from which the heads were recarved. Broken thrones are similarly identified as nothing more than corollary evidence of the actual process of recarving itself. Future applications of the same program may engender a new generation of exciting alternatives for understanding Mesoamerican remains.
