Weaving Lessons

Language, Cognition and Cultural Change in a Mayan Community in Chiapas

By Linda Abarbanell

Haga clic aquí para leer en español.

Alonso’s mother is hovering over me. Occasionally she takes over, showing by example, counting out each thread, “p’ej” (one), then “cheb… oxeb…” I want to see the whole pattern laid out before me at once, but she teaches me to weave one row at a time, hilito por hilito, strand by strand.

Figure 1. The family altar in Tenejapa.

It is nice to spend time alone with the women. They speak very little Spanish, which is good for my Tseltal, the Mayan language spoken in Tenejapa, a rural municipality in the Chiapas highlands. I have come here to study how Tseltal speakers talk about locations and directions and whether this affects their spatial cognition. That is, can the language we speak actually shape our thoughts such that speakers of different languages arrive at radically different world views? This possibility, known as the linguistic relativity or Whorfian hypothesis, has fascinated people for centuries and continues to be a source of heated debate.

The weaving lessons generate curiosity among the children that pass in and out of the family compound, playing in the open patio. They are the many grandchildren of Alonso’s mother’s eleven children. I do not address Alonso’s mother by name (I’ve used invented names for everyone in Tenejapa, including Alonso, to protect their privacy). Alonso’s mother, my teacher, is simply jme’tik, a term of respect for a woman, the matriarch of this family that I have come to consider my adoptive kin. When I first mentioned that I wanted to learn how to weave, she strapped on her backstrap loom, displaying her skills. I was reminded of the first time I saw her as a participant in one of my studies. I didn’t know at the time that she was Alonso’s mother. She sat proudly as now, back straight, head high, and eyed me suspiciously as I showed her how the turntable constructed out of Tinkertoys worked. A rubber duck was attached to a spool in the middle to mark the distinction between left and right, front and back. Four wooden spokes extended from each side with identical square lidded boxes attached to their ends. I hid a coin in the box to the left of the duck while Alonso’s mother watched. She then covered her eyes while I spun the apparatus until the duck and its boxes were in a new orientation. Alonso’s mother then opened her eyes and I asked her to find the coin. Like many of the adult Tseltal speakers that I tested, she did not succeed. Although this might seem an easy task, participants must attend to and distinguish the left from the right side of the duck and then track this distinction after its rotation. Could language play a role in developing these skills, I wondered?

Figure 2. The indigenous saints in the church in Tenejapa. The marks on the glass are from people making the sign of the cross.

Tseltal speakers do not typically use words like “left” and “right” to talk about locations and directions. Instead of using such a viewer-centered perspective or frame of reference, they prefer to use an environment-centered frame based on anything from the location of a chair or a river to the overall uphill/downhill slope of their terrain. Over two decades ago, researchers affiliated with the Max Planck Institute for Psycholinguistics, working across more than 20 languages including Tenejapa Tseltal, found a striking correlation between the frame of reference speakers habitually use in their language and their performance on certain spatial tasks. For example, participants might view a row of toy animals placed on a table before them, left-to-right or uphill-to-downhill, depending on one’s perspective. They then turn 180° and walk to another table where they are handed the animals and asked to make the “same” array. Crucially, after turning, the participants’ left and right turns with them while environment-derived axes like uphill/downhill do not. Speakers of languages like Tseltal recreated the array from an environment-centered perspective, while speakers of languages like English rotated the array with their own left/right.

One might think a left/right solution in this case would be easy since participants are using their own perspective, unlike the turntable task which requires tracking the perspective of the duck. Thinkers from Immanuel Kant to the psychologist Jean Piaget have argued that such an egocentric viewer-centered perspective is the natural result of our embodied cognition. The Max Planck researchers, however, and some still today, argued that the frame of reference speakers use in their language may actually re-structure speakers’ nonlinguistic spatial cognition. Penelope Brown and Stephen Levinson, in their 1992 paper on ‘left’ and ‘right’ in Tenejapa, proposed that the absence of a left/right reference frame in their language resulted in a “conceptual gap” in Tenejapan’s left/right conceptualization of space. More recently, researchers like Daniel Haun and colleagues have suggested that a left/right perspective might only be constructed through language, with an environment-centered perspective being our evolutionarily-inherited default state. Psychologists Peggy Li and Lila Gleitman, however, argued that an alternative explanation to the Max Planck results was also plausible: Two things can be the “same” in many different ways. The tasks were open-ended in the sense that they could be solved from either a viewer-centered or environment-centered perspective. Participants may simply have interpreted the command to “make it the same” in line with how their language community typically talks about locations and directions. That is, might previous results have been an effect of participants’ habitual language use on their interpretation of such an open-ended task?

Figure 3. The tapestry woven by the author during the weaving lessons, produced on a traditional backstrap loom.

I arrived in Tenejapa in the summer of 2004 with an arsenal of tasks designed to answer this question. Tenejapa is situated on a 61.8square miles rectangular plot of highland terrain that slopes from about 1.7 miles in the south to half a mile in the north. As you descend, you can feel the temperature rise and see the vegetation change. Years ago, many families owned land in both the cooler uphill and warmer downhill regions, spending part of each year tending their downhill crops. The uphill/downhill distinction was so prominent that it came to define the main environment-centered axis Tenejapans use for spatial reference. Even when talking about small-scale space, a Tenejapan might ask you to “pass the fork to the downhill from the bowl.” Tseltal has words for “left” (xin) and “right” (wa’el), but traditionally these were used only to label body parts and were not extended to describe regions of space. Left/right asymmetries in general were reportedly underemphasized in a community where many adults over the age of about 35 still have little formal schooling and literacy skills. Unlike front/back, left/right is a subtle asymmetry. It takes English-speaking children years to acquire these terms, and many adults still frequently confuse them. Was it plausible, then, that Tenejapans perceived their “tilted world,” as Brown and Levinson described it, in a radically different manner?

The tasks designed by Peggy Li, Anna Papafragou, Lila Gleitman and I, while based on the Max Planck tasks, made it clear which perspective we wanted participants to use, either viewer-centered or environment-centered. When comparing participants’ errors rates across two such matched conditions, we found that Tseltal speakers could readily use both perspectives and even had an easier time using a viewer-centered perspective on some tasks—at least, for tasks that relied on participants’ own egocentric perspective. More recently, Peggy Li and I have found that things like the distance between tables and the degrees participants turn can influence which perspective is easier to use across language groups. But perhaps the Tenejapa that I encountered was different from the Tenejapa that had been documented over a decade before.

Figure 4. María showing me how to finish off the edges.

Thanks to national development programs, virtually all children now attend school at least through the primary level where they learn Spanish. Meanwhile, a rapidly growing population, depleted land resources and increasing consumer needs have rendered the subsistence agriculture structure of the community difficult to sustain. Many young people have no desire to take on the roles and responsibilities of the past. Young men migrate to urban areas of Mexico and the United States and send remittances home. They return, bringing with them new goods and ways. Cultural change is slow, but nevertheless it seeps in like rain dripping through a hole in the corrugated aluminum roof above my head.

At times while I am weaving, I look over at the family altar in the corner and it suddenly looks like something ordinary that I have lived with my entire life—the faded fabric flowers and portraits of saints, the hand-hewn wooden crosses and wrought aluminum candle tray, catching the paraffin that drips quietly beneath blue flames. I am not sure, however, if I am adapting to their culture or have subsumed it within my own. Alonso and I talked of many things that first summer. He asked many questions about the United States. He told me he was not married, but I soon learned that the young woman sitting in the patio with her three small daughters was his wife. “Perhaps he was embarrassed to tell you about his life,” his younger brother, Marco said. It was two years before I was able to return to Tenejapa, and Alonso was not there. He had waited for me the following summer before going to the United States to work, so I worked with Marco instead. The next time, Marco, too, had gone. Their older sister, María, shyly asked if I would work with a woman. More than a research assistant, she has been my commadre, my friend.

Figure 5. Alonso’s mother, weaving in the family patio.

With María’s assistance, I’ve run many studies over the years. Assessments of left/right language comprehension and use have helped assuage my fears that the language has completely changed. Most adults still do not spontaneously use these terms. Children have incorporated the Spanish terms for left and right, izquierda and derecha, into their Tseltal discourse, but most are confused as to which side is which and may sometimes even extend them to front and back. I have little doubt that the language is changing, as is the culture—it’s just difficult to assess exactly how and how fast. These changes present an opportunity for research. In one study, I used expected variation in the exposure to “left” and “right” to compare three groups of 10-12 year old Tenejapan children on the turntable task at which the adults had failed: Tseltal-speaking children living in a rural region, Tseltal-speaking children living in the somewhat more urban municipal center, and native Spanish-speaking children, also in the center, drawn from the small population of mestizo families that have lived there for generations.

Figure 6. The municipal center, or lum, viewed from an outlying paraje or subdivision of Tenejapa,

Other researchers have found that it is not until English-speaking children are about 10 years old that they succeed at a similar non-egocentric left/right perspective-taking task. This, coincidentally, is about the age they begin to correctly use left/right language non-egocentrically (“pass me the fork to your left”). By testing three groups of similar-aged children expected to vary in their acquisition of left/right language, we hoped to disentangle the contribution of such language from general cognitive development in promoting their success on such perspective-taking tasks. Unlike the adults who may have been unfamiliar with the pragmatics of a school-like task, the children all attended school. The two center groups were even in the same classes and had similar grades. Our results indeed yielded the predicted variation in the children’s understanding of “left” and “right;” and, while not necessary, this understanding correlated with their performance on the turntable task. In fact, when we trained children who initially did not reach a certain criterion correct on the task to apply left/right labels to the sides and boxes of the duck, this boosted their performance above that of a control group. Spatial frames of reference in language may not radically restructure basic embodied spatial perception, but they may help speakers represent and manipulate less readily given relationships such as the left/right perspective of someone else.


Figure 7. A tradition cornfield, or milpa, and house in a rural paraje in Tenejapa.

What happens when two languages collide that have very different systems for representing relationships? Are these worlds woven together like the tapestry growing beneath my hands, its complex patterns encoding remnants of an ancient Mayan cosmology? The Tseltal-speaking children playing in the patio, watching a foreigner learn a craft many of them never will, will eventually learn their left and right and this might help them take the left/right perspective of another. But what perspective is lost? The children I tested were familiar with the Spanish terms izquierda and derecha, but few knew their Tseltal equivalents, xin and wa’el. Was the tilted world that accompanied them also gone? The last time I was in Tenejapa, Alonso had recently passed away. I had seen others there grow old and die before their time, consumed by alcohol, poverty and the alienation that comes from centuries of oppression by a more dominant culture. Alonso’s death, however, struck hard. He had been fighting with his wife one night when, drunk, he fell from the stairs in his own home. María and I, accompanied by her father, made our way to the small cemetery, or panteón so I could pay my respects. “Alonso,” their father said, pointing to the concrete slab at our feet. I nodded but felt numb. He took the candles he’d brought and welded them to the grave with their own wax. Lighting them, he began to chant in the lilting cadence of traditional Tseltal prayer. Few people know how to pray in this way. There is no guidebook, no written pattern—it is transmitted, not taught. As he prayed, the rain began to fall, mixing with my tears to stain the earth.

Figure 8. A view of a rural house and milpa, with the sacred peaks of Mukul Ajawand Ch’in Ajaw in the distance.

It took me two months to weave a tapestry 8 x 17 inches in size. It sits now, proudly, upon my shelf.


Linda Abarbanell, Ed.D. (Harvard, 2010), is an Assistant Professor of Psychology at San Diego State University, Imperial Valley. She studies the relationship between language, culture and cognition in Chiapas, Mexico and at the US-Mexico border. Her first book, The Reproductive Rights of Mayan Women: Prejudice, Poverty, and Public Policy is expected to be forthcoming next year.