![Sample our Social Sciences journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/110822/TOC-Subject-Sample-2021-Social-Sciences.jpg"})
Abstract
This study examines ethno-racial identification among urban Peruvians, with special attention to those who are at risk of ‘de-Indianizing’. Specifically, we use a nationally representative survey to describe how city residents classify themselves and how self-classifications are associated with primordial and circumstantial characteristics. Consistent with official statistics, a large majority identifies as mestizo. However, the share that self-identifies as indigenous is much smaller than expected from official language-based criteria or from appearance alone. Moreover, identification is rooted in primordial characteristics associated with descent and the family-of-origin's linguistic environment, particularly for those with indigenous ancestries. Identification is also linked to socio-economic circumstances and perceived discrimination.
Notes
1. Numerous other descriptors exist. For example, ‘campesino’ is preferred over the derogatory ‘indio’ to designate Amerindian peasants. Sometimes used pejoratively, ‘cholo’ may appeal to young, educated, indigenous-origin bilingual urbanites (de la Cadena 2000; García 2005).
2. Ambiguity is illustrated by de la Cadena's claim that Peruvians consider her white despite her physical features and ‘would laugh … if I claimed to be Indian’ (2005, p. 261). Another study revealed that some subjects describe her as mestizo (de la Cadena 2000).
3. Compared to Spanish monolinguals, indigenous language speakers have higher poverty rates (63 vs 40 per cent), less schooling (6.6 vs 8.0 years) and limited sanitation (Trivelli 2006).
4. Indigenous-language speakers are less than seven per cent of the population of the Department of Lima.
5. De-Indianization suggests urban migrants with indigenous backgrounds are more likely than rural non-migrants to identify as mestizo (Sulmont 2011). This cannot be examined here because rural residents were excluded from the survey.
6. The ENNIV was part of a World Bank research initiative. GRADE developed the supplement and Instituto Cuánto conducted the fieldwork.
7. 86 per cent of original urban households were re-interviewed. All eligible persons were interviewed in 54 per cent and a subset was interviewed in 32 per cent of households.
8. Negative experiences are not highly correlated. All inter-item correlations range from .00 to .20, except for connections and both socio-economic background (.33) and other reasons (.28).
9. Research in other Latin American settings suggest that similar visual measures based on skin tone have high reliability and are unaffected by interviewer characteristics (Villarreal 2010).
10. For example, the term ‘indigenous’ is in the lexicon of national and international agencies that focus on all Amerindians. Still, complexities are involved in its usage. On the one hand, some claim that ‘indígena historically refers primarily to Andean peoples. Its use was largely absent in Amazonia’ (Greene 2006, p. 34). On the other hand, some imply that Andean migrants might prefer specific terms instead of ‘indigenous’, which may be understood to refer to Amazonians or have negative connotations (Valdivia 2003). In our view, those preferring specific identifiers such as ‘Quechua’ would be unlikely to reject ‘indigenous’ here given that few chose ‘other’ and differences by specific language were not significant (shown below).
11. Relative-risk ratios divide the probability of an event for a group by the probability for a comparison group.
12. This is confirmed by statistical tests. Also, combining all indigenous-language speakers produces significant estimates in Models 1–3.
13. Thirteen tests for interactions were conducted (e.g., income*ancestry, education*ancestry, appearance*income, race discrimination*education, sex*income, sex*education). Only migration*region was significant. Unlike those in other regions whose pattern mirrored the main effects, rural-urban migrants are indistinguishable from non-migrants in the selva. This result is not substantively important.
14. Among those with a Spanish-monolingual mother and indigenous-language father, only one per cent self-identified as indigenous, 11 per cent as white, and 88 per cent as mestizo. Of those with two indigenous-language parents, 15 per cent self-identified as indigenous, 8 per cent as white, and 76 per cent as mestizo. The latter figure is particularly noteworthy. Figures for those with an indigenous mother and Spanish-monolingual father fall between these values.
15. Racial discrimination is unrelated to education. Place-of-origin and language-based discrimination are negatively associated with education. Education and lack of connections are positively related.
