• Adger, W. N., , Kelly P. M. , , Winkels A. , , Huy L. Q. , , and Locke C. , 2002: Migration, remittances, livelihood trajectories, and social resilience. Ambio, 31, 358366.

    • Search Google Scholar
    • Export Citation
  • Bucholtz, M., , and Hall K. , 2005: Identity and interaction: A sociocultural and linguistic approach. Discourse Stud., 7, 585614.

  • Canal Rural, cited 2010: Previsão do tempo. [Available online at http://www.canalrural.com.br/canalrural/jsp/default.jsp?uf=1%26local=1%26capa=963%26section=previsaodotempo.]

    • Search Google Scholar
    • Export Citation
  • Climatempo, cited 2010: Região nordeste. [Available online at http://www.climatempo.com.br/previsao-do-tempo/regiao/ne/nordeste.]

  • EMATERCE, 2010: Situação da safra agrícola, 31 de Julho de 2010. Ceara State Rural Extension and Technical Assistance Enterprise Rep., 34 pp. [Available online at http://www.ematerce.ce.gov.br/arquivos/SITUACAODASAFRAAGRICOLA31JUL2010.pdf.]

    • Search Google Scholar
    • Export Citation
  • Green, D., , and Raygorodetsky G. , 2010: Indigenous knowledge of a changing climate. Climatic Change, 100, 239242.

  • Henshaw, A., 2006: Winds of change: Weather knowledge amongst the Sikusilarmiut. Climate Change: Linking Traditional and Scientific Knowledge, R. Riewe and J. Oakes, Eds., Aboriginal Issues Press, 177–186.

    • Search Google Scholar
    • Export Citation
  • IBGE, cited 2010: Censo agropecuário 2006: Resultados preliminares. [Available online at http://www.ibge.gov.br/home/estatistica/economia/agropecuaria/censoagro/2006/agropecuario.pdf.]

    • Search Google Scholar
    • Export Citation
  • Ingram, K., , Roncoli C. , , and Kirshen P. , 2002: Opportunities and constraints for farmers of West Africa to use seasonal precipitation forecasts with Burkina Faso as a case study. Agric. Syst., 74, 331349.

    • Search Google Scholar
    • Export Citation
  • INMET, cited 2010: Normais climatológicas do Brasil 1961-1990. [Available online at http://www.inmet.gov.br/html/clima/mapas/?mapa=prec.]

    • Search Google Scholar
    • Export Citation
  • Krupnik, I., , and Jolly D. , Eds., 2002: The Earth is Faster Now: Indigenous Observations of Arctic Environmental Change. ARCUS, 356 pp.

  • Lemos, M. C., , Finan T. , , Fox R. , , Nelson D. , , and Tucker J. , 2002: The use of seasonal climate forecasting in policymaking: Lessons from northeast Brazil. Climatic Change, 55, 479507.

    • Search Google Scholar
    • Export Citation
  • Livingstone, D., 2010: Conversations on rain-making. Perspectives on Africa: A Reader in Culture, History and Representation, R. R. Grinker, S. C. Lubkemann, and C. B. Steiner, Eds., Wiley-Blackwell, 245–248.

    • Search Google Scholar
    • Export Citation
  • McCrea, R., , Dalgleish L. , , and Coventry W. , 2005: Encouraging use of seasonal climate forecasts by farmers. Int. J. Climatol., 25, 11271137.

    • Search Google Scholar
    • Export Citation
  • Meinke, H., , and Stone R. , 2005: Seasonal and inter-annual climate forecasting: The new tool for increasing preparedness to climate variability and change in agricultural planning and operations. Climatic Change, 70, 221253.

    • Search Google Scholar
    • Export Citation
  • Nelson, D., , and Finan T. , 2009: Praying for drought: Persistent vulnerability and the politics of patronage in Ceará, northeast Brazil. Amer. Anthropol., 111, 302316.

    • Search Google Scholar
    • Export Citation
  • Orlove, B., , Broad K. , , and Petty A. , 2004: Factors that influence the use of climate forecasts: Evidence from the 1997/98 El Niño event in Peru. Bull. Amer. Meteor. Soc., 85, 17351743.

    • Search Google Scholar
    • Export Citation
  • Orlove, B., , Roncoli C. , , Kabugo M. , , and Majugu A. , 2010: Indigenous climate knowledge in southern Uganda: The multiple components of a dynamic regional system. Climatic Change, 100, 243265.

    • Search Google Scholar
    • Export Citation
  • Patt, A., , and Gwata C. , 2002: Effective seasonal climate forecast applications: Examining constraints for subsistence farmers in Zimbabwe. Global Environ. Change, 12, 185195.

    • Search Google Scholar
    • Export Citation
  • Patt, A., , Suarez P. , , and Gwata C. , 2005: Effects of seasonal climate forecasts and participatory workshops among subsistence farmers in Zimbabwe. Proc. Natl. Acad. Sci. USA, 102, 12 62312 628.

    • Search Google Scholar
    • Export Citation
  • Pennesi, K., 2006: Aspectos lingüísticos nas previsões de chuva: FUNCEME e os profetas da chuva. Profetas da Chuva, K. P. H. Martins, Ed., Tempo d’Imagem, 144–155.

    • Search Google Scholar
    • Export Citation
  • Pennesi, K., 2007a: Improving forecast communication: Linguistic and cultural considerations. Bull. Amer. Meteor. Soc., 88, 10331044.

    • Search Google Scholar
    • Export Citation
  • Pennesi, K., 2007b: The predicament of prediction: Rain prophets and meteorologists in northeast Brazil. Ph.D. dissertation, University of Arizona, 387 pp.

    • Search Google Scholar
    • Export Citation
  • Roncoli, C., 2006: Ethnographic and participatory approaches to research on farmers’ responses to climate predictions. Climate Res., 33, 8189.

    • Search Google Scholar
    • Export Citation
  • Roncoli, C., , Orlove B. , , Kabugo M. , , and Waiswa M. , 2010: Cultural styles of participation in farmers’ discussions of seasonal climate forecasts in Uganda. Agric. Hum. Values, 28, 123138.

    • Search Google Scholar
    • Export Citation
  • Saville-Troike, M., 2003: The Ethnography of Communication: An Introduction. 3rd ed. Blackwell, 336 pp.

  • Secretaria de Agricultura e Pecuária, 2002: II plano indicativo de desenvolvimento rural do Ceará 1999-2002: Rumo ao desenvolvimento rural do Ceará. Governo do Ceará, 97 pp.

    • Search Google Scholar
    • Export Citation
  • Serviço Geológico do Brasil, 1998: Map of Quixadá. Ministério de Minas e Energia.

  • Stuart, N., and Coauthors, 2006: The future of humans in an increasingly automated forecast process. Bull. Amer. Meteor. Soc., 87, 14971502.

    • Search Google Scholar
    • Export Citation
  • Suarez, P., , and Patt A. , 2004: Cognition, caution and credibility: The risks of climate forecast application. Risk Decis. Policy, 9, 7589.

    • Search Google Scholar
    • Export Citation
  • Wilhite, D., 1993: The enigma of drought. Drought Assessment, Management and Planning: Theory and Case Studies, D. Wilhite, Ed., Kluwer Academic Publishers, 3–15.

    • Search Google Scholar
    • Export Citation
All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 120 120 5
PDF Downloads 32 32 2

Making Forecasts Meaningful: Explanations of Problematic Predictions in Northeast Brazil

View More View Less
  • 1 University of Western Ontario, London, Ontario, Canada
© Get Permissions
Full access

Abstract

This study illustrates the need to consider the multiple interpretations and experiences that influence how climate forecasts are evaluated in local contexts when assessing how useful forecasts can be for increasing the resilience of rural communities. Video clips of predictions made by scientific and traditional forecasters were shown in interviews and focus groups to elicit explanations for why the predictions are sometimes judged to be inaccurate, not useful, or inappropriately communicated by different sectors of the rural population in Ceará, Northeast Brazil. Results indicate that climate forecasts are not simply a decision-making tool that provides information in a one-way transfer from forecaster to user. The meanings and values of predictions are jointly created by both forecasters and their audiences. Predictions and the discussions that surround them are also an important part of expressing social identities and ideas about how the world works. Ineffective predictions are explained here in terms of religious beliefs, environmental change, forecaster identity, interactional context, and cultural practices.

Corresponding author address: Karen Pennesi, Assistant Professor, Department of Anthropology, Social Science Centre, University of Western Ontario, London ON N6A 5C2, Canada. E-mail: pennesi@uwo.ca

This article is included in the Ways of Knowing special collection.

Abstract

This study illustrates the need to consider the multiple interpretations and experiences that influence how climate forecasts are evaluated in local contexts when assessing how useful forecasts can be for increasing the resilience of rural communities. Video clips of predictions made by scientific and traditional forecasters were shown in interviews and focus groups to elicit explanations for why the predictions are sometimes judged to be inaccurate, not useful, or inappropriately communicated by different sectors of the rural population in Ceará, Northeast Brazil. Results indicate that climate forecasts are not simply a decision-making tool that provides information in a one-way transfer from forecaster to user. The meanings and values of predictions are jointly created by both forecasters and their audiences. Predictions and the discussions that surround them are also an important part of expressing social identities and ideas about how the world works. Ineffective predictions are explained here in terms of religious beliefs, environmental change, forecaster identity, interactional context, and cultural practices.

Corresponding author address: Karen Pennesi, Assistant Professor, Department of Anthropology, Social Science Centre, University of Western Ontario, London ON N6A 5C2, Canada. E-mail: pennesi@uwo.ca

This article is included in the Ways of Knowing special collection.

1. Introduction

As droughts become more frequent in semiarid regions, meteorological agencies and traditional forecasters gain importance as local institutions working to increase resilience of rural communities under the stress of such climatic changes. Resilience refers to “the ability of communities to absorb external changes and stresses while maintaining the sustainability of their livelihoods” (Adger et al. 2002, p. 358). Among weather and climate forecasters, a common assumption is that the purpose of forecasting is to provide timely, reliable, and comprehensible information to users (i.e., individuals, institutions, government) to help them make strategic decisions that enable successful adaption to variations, changes, and stresses. This assumption leads scientific forecasters to identify various “users” of their “products,” and efforts are made to improve the type, accuracy, accessibility, and understanding of the information in accordance with user’s needs and forecasting capabilities (McCrea et al. 2005; Stuart et al. 2006). From this perspective, predictions are seen as instruments of knowledge that are transferred from expert forecasters to nonexpert users, increasing the user’s capacity to make optimal choices and sustain livelihoods, even when weather- and climate-related hazards occur.

There are several complicating factors that prevent the knowledge transfer and application approach from working successfully. Researchers attempting to address these problems have investigated impediments for small-scale agricultural producers to use forecast information (Lemos et al. 2002; Orlove et al. 2004; see special issue of Climatic Research, 2006, Vol. 33, No. 1; Suarez and Patt 2004). Initially, lack of access to the forecasts and limited understanding of scientific information was identified as a common problem. In response, several studies done with farmers in Africa suggest that participation in workshops or discussion groups increases comprehension and use of scientific forecasts in agricultural decision making (Patt et al. 2005; Roncoli et al. 2010). Next, research on decision making has shown that forecasts are only one type of information in a complex process of determining which actions will be taken (Meinke and Stone 2005; Roncoli 2006). It is, therefore, difficult to assess the impact and utility of forecasts in particular contexts if the assessment is focused only on planting strategies and crop yields. Further complicating the situation in some cases is the existence of local prediction practices that offer complementary or competing sources of knowledge. In Uganda, for example, farmers are both producers and consumers of indigenous climate forecasts (Orlove et al. 2010), and scientific forecasts are being incorporated into this arena. Recent work examining attempts to integrate indigenous or traditional knowledge with scientific forecasting shows that there is a wide variation in the level of integration and ultimate success of this approach in increasing the resilience of vulnerable populations (Green and Raygorodetsky 2010). Studies of vulnerability done in the last two decades have expanded our knowledge beyond earlier work to include sociopolitical relations of power, access to entitlements, and agency as important factors influencing a population’s adaptive capacity when faced with environmental or other hazards (Nelson and Finan 2009). These factors effectively limit which alternatives are available to forecast users who wish to make decisions based on the information.

Finally, as this study demonstrates, users do not always interpret predictions as forecasters intend, nor do they judge a prediction’s effectiveness according to the forecasters’ criteria. Reporting on research from Ceará, Northeast Brazil, this paper contributes to our understanding of how farmers interpret and evaluate climate forecasts, offering explanations for why some predictions are judged by them to be inaccurate, not useful, or inappropriately communicated. I argue that religious beliefs and the social identities of forecasters are important factors in this assessment, in addition to subjective evaluations of accuracy and varying levels of scientific understanding. Examples of traditional forecasts are presented here, in comparison with scientific forecasts, to demonstrate how farmers use the same criteria for evaluating their effectiveness and to illustrate the variety of prediction styles they encounter.

Evaluating the effectiveness of forecasts in increasing the resilience of particular communities requires a close look at the multiple meanings of forecasts in local cultural contexts. The findings discussed here challenge the assumption that the value of forecasts is restricted to their utility in decision making and deepen our understanding of the cultural meanings of prediction. The Ceará case shows that predictions can contribute to community resilience by reinforcing community bonds and promoting the value of sustaining agricultural livelihoods.

a. Context of prediction in Ceará, Brazil

In Ceará, a state in Northeast Brazil, climate forecasts are greatly anticipated and taken seriously in everyday conversation. Agricultural livelihoods occupy 80% of rural workers in Ceará and 35% of its nearly seven million people live in rural areas (Secretaria de Agricultura e Pecuária 2002). Almost all of the agricultural establishments in Ceará are small holdings that rely on the labor of family members. More than half of the households engaged in family farming to supplement their meager income with government pensions and benefits (IBGE 2010). Expensive irrigation systems are used in a few small regions, but most of the corn, beans, rice, and vegetables grown for subsistence are dependent on seasonal rainfall. Ninety-five percent of Ceará has a semiarid climate, with average temperatures of 26°C year-round, and most of the annual rainfall normally occurs in the months from February to May (INMET 2010). Typically, food crops and pasture grasses are planted in January, when soils have become sufficiently moist with what meteorologists call “preseason” rains. The expectation is that adequate rains will recur with some regularity over the next few months for the plants to grow well and produce a harvest. If the crops fail to mature because of flooding or dry spells at the beginning of the season, then fields may be replanted in March or April to try to take advantage of the second half of the year. Irregularly distributed rains are characteristic of semiarid climates and agricultural droughts, in which harvests are well below average, are common. In the last two decades, agricultural droughts have increased in frequency, happening every 3–5 yr (J. M. B. Alves 2010, personal communication). These conditions lead to an undercurrent of anxiety at the beginning of each rainy season, which becomes more palpable if the weeks pass and there is little rain.

There is a keen interest in both climate and weather predictions among Ceará’s rural population. People want to know whether there will be sufficient rains for growing crops over the entire season as well as when dry periods may be expected and how long these will last. This information is valuable in part because several choices can be made depending on the expected amount and distribution of rainfall. Farmers can decide to plant proportionally more or less of particular crops (e.g., corn requires more water than beans) or select seed varieties that grow faster or produce higher yields. They can choose to cultivate higher or lower areas to catch runoff or avoid flooding. They can seek credit to plant a larger area, or they can decide that planting is not worthwhile and seek other means of income, including off-farm work or government assistance. In addition to farmers and livestock producers, business owners and service providers who depend on the farmers as their main customers also seek information that will help them make planning decisions related to their products and the likely spending habits of their clientele.

Predictions are sought from a limited number of sources. “Traditional” forecasters are known as profetas da chuva1 (“rain prophets”). For this paper, I follow local language usage in Ceará and use “traditional” to refer to the kinds of predictions made by rain prophets as distinct from the “scientific” predictions made by meteorologists. The tradition of prediction associated with rain prophets involves the development of a knowledge system similar to that of the Ugandan system described by Orlove et al. (2010): knowledge of the seasonal patterns of precipitation and temperature; a set of local climate indicators based on the behavior of insects, animals, birds, and plants; observation of meteorological events; and information about the development of the rainy season in other regions. Rain prophets use no equipment but rely on their own five senses and sometimes on information gathered from others to make their predictions. An essential component of this tradition, as will be discussed below, is that there must be some personal connection between the rain prophet and the land, either through experience or by learning from someone with such experience. Many rain prophets are farmers but some in Quixadá, where fieldwork was conducted, have other occupations, including a dentist, an accountant, a teacher, and a restaurant owner. Rain prophets are consulted during day-to-day interactions in their communities as part of casual conversation, and many of them participate in the annual meeting of rain prophets held in the center of Quixadá each January.

For those interested in scientific predictions, there are news broadcasts and other television programs that disseminate forecasts from state and federal meteorological institutions, including the Ceará Agency for Meteorology and Water Resources (FUNCEME), the National Meteorology Institute (INMET), and the Center for Weather Prediction and Climate Studies (CPTEC) within the National Institute for Space Research (INPE). There are also Web sites that provide local forecasts, such as Climatempo (2010) and Canal Rural (2010). The most commonly discussed forecast for the rainy season in Ceará is FUNCEME’s climate forecast for February–May, which is the result of an international workshop held in late January each year.

Among farmers, there is widespread awareness of various predictions for each year, which they combine with their own experience, local knowledge, and traditional agricultural practices to form an opinion about how the season is unfolding. While research shows that farmers in Ceará do not explicitly attribute agricultural decisions to seasonal forecasts (Lemos et al. 2002; Pennesi 2007b), predictions are important in creating a sense of community, as they provide focal points around which people discuss their plans, worries, and achievements. Staying informed about particular forecasts enables one to participate in such conversations, offering opinions in support or disagreement with one or another forecaster, and reporting on the accuracy of the prediction in reference to locally experienced conditions. Many people describe following the various predictions in terms of a game or sporting event; people like to see who will get it right or wrong and it is entertaining. This is similar to the situation in Zimbabwe, where farmers “felt that the forecasts improved the quality of their lives, even if the ways in which they did so were not concrete and identifiable” (Patt and Gwata 2002, p. 193). The strong social bonds that predictions reinforce make the community more resilient, in that social networks provide access to resources that can be used in response to environmental stresses such as drought.

b. Religious beliefs, social identities, and predictions in rural Ceará

This research demonstrates that understandings about how prediction works are closely tied to religious beliefs. I offer here a brief summary of these beliefs based on ethnographic data gathered over the last 8 yr.

Among the rural population of Ceará, there is a widely shared belief that God is all powerful and the ultimate determiner of how much rain will fall, when, and where. Claiming absolute certainty in predictions is considered both impossible and foolish because God can change things instantly and inexplicably. The best course of action, both morally and practically, is to pay close attention to nature because it is within the processes of nature (God’s creation) that God’s plan is revealed, if one is able to correctly interpret the signs. Within this belief system, it is necessary that someone attempting to make a prediction of rain (i.e., God’s actions) be properly reverent and demonstrate faith so that observers will trust that the forecaster is prepared to receive God’s message, rather than daring to circumvent this God-given knowledge and devise his own human-centered knowledge. Thus, rain prophets may paradoxically claim that they are confident in their prediction because they have faith and have carefully observed the signs in nature, while adding a disclaimer of uncertainty because God is the only one who knows for sure what will happen. Such reasoning, in which God’s will and actions are considered in conjunction with human actions, is encountered frequently in anthropological literature. In one often-cited example, Dr. David Livingstone presents a conversation between a medical doctor and a rain doctor in South Africa, demonstrating that both use their knowledge and medicines to help their people, even as they recognize the limitations of their own system and the ultimate power of God to decide outcomes (Livingstone 2010, 247–248).

Most study participants say they are Christian (Catholic or Evangelical), and the beliefs described above are based on an oral tradition of teachings attributed to the Bible, shared by the largely illiterate older rural population. Despite the prevalence of religious references in discourse, the local Catholic bishop reports that only 10% of Quixadá’s Catholics are baptized, noting that very few go to church. The bishop does not see the rain prophets as a threat to the Church because they are not making any miraculous claims and they do promote faith in God. Counter to what some rain prophets claim, the bishop did not support the idea that rain prophets had a privileged relationship with the divine, explaining that every human talent is a gift from God.

Religious discourse is only one arena in which identities are created and challenged. Current discussions in anthropological literature treat identity not as a list of traits that an individual or group possesses but as an ongoing performance of actions and attitudes that are made meaningful through the interaction of individuals with others. Bucholtz and Hall (2005) propose a framework for analyzing identity within linguistic interactions based on the following principles: 1) identity emerges as the product of interaction; 2) identities encompass demographic categories, roles particular to an interaction, and culturally defined roles; 3) identities may be referred to directly by labels or indirectly by implication or symbol; 4) identities are constructed through various relationships between self and other; and 5) identities may be intentional, habitual, conscious, or imposed. In the context of prediction, identities, such as “rain prophet,” “farmer,” “forecast user,” or “expert,” are shaped by the participants in the interaction and the type of interaction itself, by what is said or not said, how people do or do not act, and by conscious and unconscious ideas about how one wants to be perceived. This theoretical background informs the analysis that follows.

2. Study design

Fieldwork for this study was conducted over a 6-week period in May and June 2010, in five districts of the município2 of Quixadá, in Ceará (Fig. 1). Quixadá was chosen to build on previous research (Pennesi 2006, 2007a) and because of its unique characteristics. Quixadá holds an annual meeting of the rain prophets each January and arguably has the highest concentration of rain prophets compared to other municípios in Ceará. Thus, it is a rich site for studying interpretations and evaluations of traditional predictions, in comparison to the scientific forecasts available largely through the media. Located approximately 160 km inland from the state capital of Fortaleza, Quixadá comprises 11 districts. Those included in the study were Custódio, Juatama, Tapuiará, and Cipó dos Anjos, as well as the urban center of Quixadá (Fig. 2). These districts were selected for their accessibility in relatively close proximity to the center while still being distributed throughout the município to capture variations in microclimate, as well as for the existence of active rain prophets in each.

Fig. 1.
Fig. 1.

The município of Quixadá in Ceará state, Brazil.

Citation: Weather, Climate, and Society 3, 2; 10.1175/WCAS-D-10-05005.1

Fig. 2.
Fig. 2.

Location of the study districts in Quixadá (adapted from Serviço Geológico do Brasil 1998).

Citation: Weather, Climate, and Society 3, 2; 10.1175/WCAS-D-10-05005.1

Interviews were recorded in Portuguese with 67 participants (57 men and 10 women), including: farmers, small business owners who sell products related to agriculture and livestock, community leaders, local government officers, and rain prophets. Five forecasters from FUNCEME were also interviewed in Fortaleza. Focus groups with farmers were conducted in each district, with 3–10 participants in each. The population of farmers for this research included those actively engaged in agricultural and/or livestock production, using less than 10 ha of land, and without the benefit of irrigation. All rain prophets and forecasters who participated were men because the FUNCEME personnel involved in climate forecasting were all men and the only female rain prophet in Quixadá was unavailable during the study period. Six women participated in the farmer focus groups. Focus group participants ranged in age from mid-40s to more than 70 yr. FUNCEME forecasters were younger than 50 yr old, while the rain prophets were all older than 60 yr.

After an initial discussion of forecasts that were remembered from the beginning of the season, interview and focus group participants were shown video clips of three rain prophets from Quixadá making their predictions at the annual meeting in January 2010 and a news broadcast reporting on FUNCEME’s January forecast. The three rain prophets featured in the video clips were selected based on differences in form and content of the predictions, which are described in section 3. Since there was no available video of a meteorologist making a prediction in an analogous context, the news clip was chosen because the reporter quoted FUNCEME’s official forecast and it included a meteorologist explaining the main factors that indicated a higher probability of drought in 2010. In fact, farmers rarely speak directly to meteorologists and instead hear the forecasts through the media reports like this one. The four clips ranged from 1 ½ to 4 min. The entire speech of each rain prophet was shown, and the news report was edited to only show the segment about the forecast. Transcripts of the video clips are provided in the appendixes.

The video activity was used to elicit interpretations and evaluations of the specific predictions, as well as conversations about predictions and forecasters in general. Since the rainy season was coming to an end at the time of the interviews, participants were able to make judgments about the accuracy and appropriateness of the predictions in terms of observed rainfall and agricultural production. Showing the video clips was an effective way to get people’s perceptions and reactions to predictions made 6 months before, which they may not have heard or recalled. For example, the data from the opening discussions in which we asked people to evaluate predictions they remembered indicate that only a few forecasts created lasting impressions and that these varied with individuals, depending on which ones they had heard. It is impossible within a research context to obtain data on each participant’s interpretation and evaluation of every prediction he or she hears throughout the season. The study was thus designed to simulate the kind of evaluative conversations that take place throughout the season and as it ends, in which farmers compare their production and experiences with what was expected by others. Although such conversations likely occur in smaller groups or between different sets of individuals than what was arranged for the research, the interpretive frameworks and evaluative criteria used in these conversations are assumed to be essentially the same.

Transcripts from the recorded interviews and focus groups were systematically analyzed using an ethnography of communication approach (Saville-Troike 2003), which examines the interactional context of predictions as speech events. Attention is given to the elements and criteria needed to make an appropriate and effective prediction, according to cultural and community norms, and to the process by which both speakers and listeners create the meaning of a prediction, which sometimes results in divergent interpretations. This ethnographic approach also provides a framework for exploring the ways in which speaker identities and religious beliefs are important in evaluating the appropriateness of predictions.

3. Analysis

a. Characterizing the 2010 rainy season

Everyone agrees that 2010 was a dry year in Quixadá. There is some disagreement, however, about whether it can be called a seca (“drought”) or a seca verde (“green drought”), or just an inverno fraco (“weak rainy season”). Descriptions vary depending on whether one considers rainfall amounts or crop yields, and on personal experiences and attitudes. For example, inverno fraco generally means that there was enough rain at crucial times for most rain-fed farmers to produce a small harvest and perhaps to accumulate some water in reservoirs. Using Wilhite’s (1993) typology, this would not be considered a meteorological drought because the total amount of rainfall over the season may still be within the average range. Seca verde describes the situation in which there was enough rainfall to make the vegetation look green in the landscape, but it was so little and poorly distributed that food crops did not produce a significant yield. This corresponds to an agricultural drought in Wilhite’s terms because there is crop loss. Finally, seca, as used in the community I studied, is reserved for the extreme situation in which very little rain fell over the season, with long periods (20 days or more) of dry days between the rainfall events. In this case, there is no harvest at all, and many people will not have even planted seeds if there was never sufficient soil moisture to justify an attempt. Seca includes both meteorological and agricultural droughts and often means socioeconomic drought as well, when households rely heavily on government or other external assistance to obtain sufficient food and water.

Meteorologists at FUNCEME report that in 2010, Ceará received 55% less rainfall than average during the February–May period, with a total of 435.5 mm. This makes it the second driest year in the last 20 yr. (Fig. 3) One meteorologist said that 2010 was 1 of the 5 worst droughts in the history of Ceará in terms of the deadly combination of a lack of rain and poor distribution. The season began late and then there were stretches of 30–40 days with no rain at all in many areas. In Quixadá, the situation was particularly bad, as the total rainfall was 30%–55% less than average in every district and there were long dry spells at crucial times for growing crops to develop. For example, in the district of Tapuiará, 57% of the total rainfall came in one 4-day period. The state agricultural extension agency [Ceara State Rural Extension and Technical Assistance Enterprise (EMATERCE)] reports that the soil in Ceará was not moist enough to plant before March and by the second half of March, 30% of the fields still had not been planted. With the lack of rain, poor distribution, and the failure to plant large areas, Quixadá lost 70%–90% of the expected harvest of cotton, beans, and corn for 2010 (EMATERCE 2010), making it a drought in both meteorological and agricultural terms.

Fig. 3.
Fig. 3.

Total January–May rainfall for Ceará, 1990–2010 (source: FUNCEME).

Citation: Weather, Climate, and Society 3, 2; 10.1175/WCAS-D-10-05005.1

When asked to describe the rainy season for 2010, farmers responded according to observations of rainfall and agricultural yields. In Juatama and Tapuiará, some said there had not been a rainy season (a euphemism for drought) and many had not even planted. Those who planted lost entire crops, and the associations of agricultural producers had already begun mobilizing to demand government aid.

Juatama

Farmer 1: It didn’t rain at all.

Community leader: Here it’s dry. Here there is nothing. You can go around to the houses of these poor people, if they don’t want to show themselves. But nobody has anything. I say this because I know them well, each one … How can we say that there is a rainy season?

Tapuiará

Farmer: A year of great drought here for us in our region.

Some participants reported which parts of particular months had rain for given locations; however, despite these rainfall events, the poor distribution of the rains that caused crop failures and the low total rainfall still made the season “dry” or “weak.”

Tapuiará

Farmer 1: In March not a drop. There was a rain in April. It rained well for 10 days or was it 15?

Woman: There was a good rain. There was one time that it filled that pond and stopped.

Farmer 2: It only rained 4 days.

Woman: Right. It was little rain. It all came at once and then it stopped.

Quixadá

Businesswoman (sells feed, seed, vaccines): It’s dry, there was no rainy season. There was some passing rains here and there. It’s a weak little season.

In Custódio, in contrast, the rainy season was described in more positive terms by the farmers in the focus group.

Custódio

Interviewer: Can we say this is a year of drought?

Farmer 1: It is not.

Farmer 2: For us here no.

Interviewer: But you don’t think this year is a drought year?

Farmer 3: No, I don’t think so.

Farmer 2: There can’t be a drought any more [because there had already been enough rain].

Farmer 1: Maybe another time, but this here is not.

The following sections will demonstrate how different characterizations of the rainy season had implications for how the predictions shown in the videos were interpreted and whether they were positively or negatively evaluated.

b. Interpreting and evaluating predictions

An important finding of this research is that there are multiple interpretations for any prediction, despite attempts by forecasters to restrict meanings and explain limitations. People listen for information that is relevant to them and make sense of the prediction according to their own language usage and subjective experiences. My data indicate that differences between what forecasters intend to convey and what their audiences take a prediction to mean are common, and these differences can be a source of frustration for forecasters. This study explores some of the reasons for such divergences within the rural population of Quixadá.

We now examine the four recorded predictions that were heard by study participants. For each prediction, I discuss the ways it was interpreted by various listeners, positive and negative evaluations, listener explanations for the problematic aspects, and how the forecasters evaluated their own forecast. The first two predictions were generally well received, while the second two were criticized by listeners.

1) An effective and successful prediction: Roberto

The first prediction from the annual meeting of the rain prophets, made by Roberto3 from the district of Tapuiará, lasted about 1½ min. Within the longer prediction speech (see appendix A), the key sections in which he explicitly forecasts the amount and timing of rainfall are reproduced below:

Roberto: The predictions this year, they are various4

it indicates little rains

it, a rainy season with a lot of dry spells

there’s rains, they stop and they continue to rain and a short rainy season …

little rains

at the beginning little rain

there will be a dry spell

yeah, that way

it’s not going to be a rainy season with a lot of rain

The words in bold were the ones repeated by the study participants who watched the videos when giving their interpretations and evaluations of Roberto’s forecast. In all focus groups and interviews, there was conceptual agreement with Roberto’s descriptions of how he expected the season to unfold. This means that there was recognition among the participants that a prediction had indeed been made and that there appeared to be a shared understanding of what characteristics the coming rainy season would have according to that prediction. For example, one farmer from Juatama paraphrased Roberto’s prediction as: “Little rain in January, a rain [event], some spaces, some dry spells.”

The evaluation of a given prediction depends in part on what conditions were observed and the outcome of climate-sensitive activities, such as agriculture. All of the farmers who heard the prediction indicated that they thought Roberto had made a good and accurate prediction, using the word acertou (“he got it right”) or a synonymous expression. In the comments that followed the interpretations, Roberto’s prediction was positively evaluated for its accessible language and because he brought in a tree branch with some snails on it as an example of one of the signs he observed in support of his prediction of a dry season. Roberto’s prediction elicited only one negative evaluation, which was based on the message being communicated rather than its accuracy: one farmer from Custódio said that such a negative prediction would scare him or discourage him if he had heard it in January. This is the reaction Roberto’s pessimistic predictions usually elicit at the annual meeting and was one of the reasons it was chosen for this study. In fact, the general preference among study participants for Roberto’s prediction was surprising. Overall, Roberto’s prediction was viewed favorably, and when asked which of the three predictions by rain prophets people preferred, Roberto’s was chosen in every case because it was accurate and lacked problematic aspects contained in others (discussed below). In a separate interview, Roberto judged that his 2010 prediction had been correct and was aware that others liked his performance at the meeting, reporting that there was great interest from the media and the audience members in the tree branch he had brought to show them.

2) Conceptual disagreement and religious belief: FUNCEME

FUNCEME’s forecast, as reported in the televised broadcast, was also evaluated positively, despite problems with interpretation. The news segment about the forecast was 1 min and 40 s in length, with a brief quote from a FUNCEME meteorologist explaining the influence of El Niño on the formation of rain clouds. The reporter repeated the official forecast, given here:

Reporter: The prediction is for irregular rainfall for the next three months. The estimate is around 45% chance of a rainy season below the annual historical average here in the state.

The expressions “45%” and “below the average” were the ones best remembered and repeated by study participants in their interpretations of the forecast; however, there was not a match in conceptual meaning between the meteorological interpretation and those of the Quixadá listeners. Meteorologists explain that the forecast is a probability referring to the likelihood that the total rainfall for the entire state during the February–May rainy season will be in the “below average” category. The other two possible categories are “around the average” or “above the average.” The average is the historical average of total millimeters of rain for the state, which is based on the most recent 30-yr period. This information is rarely given in news broadcasts.

In this study, farmers most commonly interpreted the forecast to mean that FUNCEME expected 45% less rainfall than the average for the season. They understood average to mean the average number of millimeters of rain either for Ceará or for Quixadá. The FUNCEME forecast, while not interpreted in the same way as meteorologists intended, still conveyed a meaning of a relatively dry year. Based on their own understandings, farmers in Tapuiará, Juatama, and Cipó dos Anjos, as well as the business owners in downtown Quixadá, all judged that FUNCEME’s forecast had been correct. They said that FUNCEME had predicted a below-average year with little rain and in fact, that is what happened. Only in Custódio was there a different opinion.

In Custódio, there was some confusion about the term below average. The farmers in the focus group explained that they did not require a lot of rain to grow crops, so even if the rainfall was below average, that did not necessarily mean it would be a drought in terms of lost agricultural production. Some knew that the average for Ceará is about 600 mm and calculated that 45% less than this would be about 310 mm, which they thought would be sufficient if it was evenly distributed. Others were confused about what the average referred to in the forecast. For instance, one farmer thought it meant an average of 300 mm per rainfall event. There had been some rainfall in Custódio, and the corn and bean plants were still growing at the time of the focus group on 15 May. The farmers expected more rain to come and that they would get a modest harvest. For this reason, they agreed that the FUNCEME forecast was incorrect because they interpreted it as predicting a dry year when they had observed rain in their district and there was still hope for a harvest. Had this focus group taken place at a later date when it was clear that crops had indeed failed, their assessment may have been different.

Another factor that may have contributed to the opinions in Custódio being different from the other districts is that a rain prophet, Inácio, was included in this group. The video clip of Inácio’s prediction was one of those shown to the other study participants but was not shown to the Custódio group, so that they would not be in the awkward position of having to evaluate his prediction with him present. Nonetheless, the influence of his optimistic outlook on the rainy season and his widely known criticism of FUNCEME was very likely a factor in the group’s evaluations. For this reason, in the focus groups that followed, we made sure no rain prophets were present.

In Juatama, Custódio, and Cipó dos Anjos, people emphasized that FUNCEME could not be completely trusted because, ultimately, they had to place their faith in God. Making such statements was a demonstration of their own proper humility and reverence, and it fit into their belief that humans are fallible beings who are expected to make errors. They said that between trusting in God and trusting in people, one must always trust God. An often-quoted popular saying warns that when humans dare to know more than God, God will change things to prove his power. A farmer from Cipó dos Anjos explains perceived errors in FUNCEME’s past forecasts in this way:

Farmer: Because God said that when man wants to know more than him

he would change things

So then that happened and God changed things

They [FUNCEME] say that a certain month will rain and it does not rain at all

But if God wants, it will rain now, here, this instant.

The idea that weather is always changing and is inherently unpredictable, despite forecasters’ best efforts, is part of the religious belief system outlined earlier. Sudden and significant changes that happen without apparent reason are explained in terms of God’s will. This means that both educated scientists and uneducated farmers are equally likely to make inaccurate predictions, and the forecasters are not held responsible. It also gives some insight as to why some farmers report that they do not make decisions based on forecasts: even if the forecast seems to be reliable, God can change things.

When asked to choose among all four predictions in the videos, FUNCEME’s and Roberto’s were the two that people preferred. In Tapuiará and among the business owners, FUNCEME was chosen over Roberto when participants were asked to decide whose prediction was more reliable and more interesting. The reason given was that FUNCEME’s prediction was scientific and based on technology. This forced choice was one of the few occasions on which there was a distinction made about the reliability of the various methods forecasters used. There is a minority opinion that claims categorically that none of the rain prophets knows anything. More often, however, there were overt statements that each forecaster, whether traditional or scientific, should be respected for his unique perspective.

3) Forecaster legitimacy and environmental change: Pedro

The second rain prophet shown was Pedro, from the urban center of Quixadá. Pedro’s prediction was the longest at 3 ½ min. After greeting and thanking the authorities, he described some of the signs he had observed and gave his forecast for 2010. He ended the prediction by reciting a poem about the joão-de-barro bird, whose nest is an important rain predictor. The section in which he clearly stated how much rain would fall and when came about halfway through and is given in the lines below.

Pedro: So it indicated an early rainy season

January will rain well

February will rain regular, a little dry spell

and but this is common

now, after the 8th of March

8th of March

that God who is our father creator and omnipotent

he will order St. Peter to turn on the shower

for all of the entire northeast of Brazil

as it is in the world

no corner will lack water

a very controlled rainy season, for sure

and whoever plants will have a healthy profit.

Study participants hearing this prediction focused on the date he specified and understood that Pedro expected plenty of rain after 8 March. One business owner summarized the forecast in this way: “He said there will be abundance,” referring to the line “whoever plants will have a healthy profit.” In contrast to this optimistic prediction, farmers had observed very little rain in March and afterward, and much of the crops had been lost. Pedro’s forecast was therefore judged to be incorrect. Even in Custódio, where farmers were still hopeful at the time of the focus group, the prediction was evaluated negatively because the rainfall had not been as abundant as they interpreted the prediction to suggest. One criticism was that he should not have been so precise about the date because that increases the chance of error and also exhibits arrogance in claiming to know something commonly understood as being impossible to know exactly. Negative evaluations included the following: “he got it wrong,” “he missed it,” “he fell down,” and “he’s a liar.”

In a follow-up interview, Pedro admitted that he had been wrong. He described how nature had “tripped him up” by showing him some signs that indicated rain, such as the joão-de-barro bird’s nest with the opening toward the west. When it did not actually rain, he said he ended up being “a liar.” The influence of the religious belief system is evident here if we interpret Pedro’s personification of nature as a reference to God. Pedro explained that for nature (God) to reveal what is happening, a person has to concentrate and have the right conhecimento intelectual (“intellectual knowledge”). Such preparation is necessary for nature (God) to “see what people need.” Since he believes that nature (God) gives people exactly what they need, Pedro blames himself for not paying proper attention or knowing how to read the signs. By making a public statement that turned out to be false, whether intentionally or not, he was “a liar” (Pennesi 2007b).

Pedro took responsibility for having misinterpreted the signs in nature, but there is a widely held belief among farmers and rain prophets that forecasting is becoming increasingly difficult because of changes in environmental or climatic conditions, which sometimes make the observed signs unreliable and lead to inaccurate predictions. This includes temporary changes as well as longer term, more persistent changes. For example, the rain prophet Roberto reported that nature is confusing things nowadays and the ants, termites, plants, and trees are getting confused. He gave the example of 2009, when nature exhibited signs of drought and then a very wet rainy season followed. Another farmer also reported that termites were giving false indications, creating wings when there was only a light mist instead of a proper rain or shedding the wings once they had already developed, leading rain prophets to make interpretive mistakes. One rain prophet blamed global warming for the appearance of contradictory signs in nature. Degradation of the environment due to human influence was also cited as a reason for a changing environment and unreliable predictors, such as particular trees that flower at unexpected times.

This kind of explanation for the reduced reliability of traditional climate predictors is also heard among indigenous peoples in Africa (Ingram et al. 2002) and the Arctic (Henshaw 2006, Krupnik and Jolly 2002). Studies such as these suggest that even as traditional and indigenous knowledge is gaining legitimacy among scientists and policy makers who are beginning to recognize the potential for integrating this knowledge with science, the knowledge producers are having to rethink their interpretive framework to accommodate a changing environment. This can have implications for their identity as prediction authorities if they make several “incorrect” forecasts before refining their knowledge of the new conditions.

Generally, however, rain prophets are not held responsible for their inaccurate forecasts when they can convincingly claim that environmental changes are to blame. In religious terms, if God is changing things in nature, humans cannot be expected to figure it out immediately. In Pedro’s case, some people did attribute his inaccurate prediction to changes in the environment but other listeners had a different explanation for his problematic forecast.

Some rain prophets and business owners categorized Pedro among the “false prophets” who go to the annual meeting seeking attention and prestige, sometimes simply repeating what they hear from other rain prophets. Since the media started reporting on the event, there has been a steady increase in the number of rain prophets making public predictions, from 6 in the first year (1997) to nearly 30 in 2011. Pedro and some of the others are not accepted as legitimate rain prophets because they have not acquired their prediction skills from time spent out on the land and consequently, they do not have the same kind of relationship with the land that the rain prophets are seen to represent. For example, Pedro has always lived in town and hosts a local radio show. He is known as a poet, and his talent in this area was exhibited in the prediction and recognized by the study participants. People said that he is a poet rather than a rain prophet; he just likes to perform. The inclusion of the poem in the prediction was not considered problematic, and comments indicated that people enjoyed it because it livened things up. One farmer said that Pedro’s focus on his performance and the poem was what had distracted him from making a proper forecast. Overall, study participants appreciated Pedro’s optimistic expressions and his attempt to be encouraging, but they found his prediction to be lacking in accuracy and credibility. The problematic aspects of his prediction resulted from his inexperience in both generating and communicating forecasts, as well as from environmental changes that affected the predictors he observed.

4) Performance and identity: inácio

The least preferred prediction was the one given by Inácio of Custódio. It lasted 1 min and 47 s, but he did not give a very specific forecast, saying only that the rainy season had already begun in January and that there would be rain in 3 months. These lines are given below:

Inácio: the rainy season has already begun …

now the rainy season is this here and there

the whole month to March

the 3 months will rain

The rest of the prediction speech concerned the rainy season in general, the abundance of rain in 2009, the supremacy of God, his own faith in God, and paying respects to the authorities and meeting organizers. None of the study participants paraphrased his forecast, as they did with the others. Responses to the question about what Inácio had said indicated that he had not made an identifiable prediction. This reaction is illustrated in the following dialogue:

Restaurant owner (sells local meat): To me he didn’t say anything, to me

Interviewer: No?

Restaurant owner: No, he confused things, what he did was more flattering one thing and another. In fact he didn’t bring anything concrete there, from what I saw, like something solid.

When asked whether Inácio’s prediction was correct or well made, people said he was wrong or that he had failed; thus, failing to make a recognizable prediction contributed to a negative evaluation of the forecast.

Study participants commented more on Inácio’s assertion of his faith in God than on his rain forecast. His deference to God and the many references to God’s power in relation to the powerlessness of humans were not considered inappropriate, since they reflected attitudes and beliefs widely shared among farmers and contributed to his credibility as a faithful observer of nature. For example, the audience applauded immediately after Inácio’s statement that his “path is God” (see appendix A). In other words, the problem was not that Inácio included the parts about God and the authorities, but that he was not clear enough about the forecast. The exchange below, between a man and a woman who each own businesses selling goods related to agriculture and livestock production, illustrates this point, as well as the way interpretations are jointly produced. The woman begins by giving her evaluation that Inácio’s speech primarily demonstrates his faith in God and suggests that he was further distracted because of his extensive flattery of the authorities. The man agrees, then extends this with the observation that as a result, he “didn’t talk about his prediction.” The woman next summarizes and rephrases the criticism that Inácio’s prediction was lacking because “He only talked about his faith in God.” Finally, the man ratifies the assessment by repeating it and the woman then concludes with a generalizing explanation that Inácio is like other farmers who plant believing that God will provide for them. Note that by using “them,” the woman signals that she does not belong to the group of farmers and rain prophets:

Woman: Yeah, that one there he’s just got faith in God because the majority of them here don’t lose their faith. But he got too excited with the authorities …

Man: Yeah, because he didn’t talk about his prediction.

Woman: He only talked about his faith in God.

Man: He only talked about his faith in God, there you go.

Woman: It’s because they really plant with faith in God, most of them.

Explanations of the problems with Inácio’s prediction were related to the performative aspects of his speech and not to his credibility or the legitimacy of his methods. For example, one farmer complained that Inácio did not talk about any of the signs he had observed as a basis for his prediction. A key point here is that Inácio is one of the most well-known and respected rain prophets, having participated in the annual meeting since its inception and having been featured in many media reports about the event. His recognized identity as a rain prophet led people to assume that Inácio had made observations and did have a forecast to present; the problem was that he did not express it well on that occasion. This contrasts with the judgments that Pedro was not a legitimate rain prophet and, therefore, criticisms of his prediction were made at a more fundamental level. Pedro’s was not discounted as simply a one-time failure because his credibility was dubious on any occasion, according to his critics. This demonstrates how forecaster identity influences the evaluation of predictions and how perceived problems are contextually framed.

Another farmer in Juatama also criticized Inácio’s exaggerated flattery of the local politicians and the meeting’s organizers, saying that he seemed to be more interested in presenting a political image than an identity as a rain prophet. While it is usual for the rain prophets to offer greetings or acknowledgments of the local authorities during their prediction speeches, Inácio made specific compliments about individuals and seemed to attribute to them the superlative achievement of making Quixadá the most respected place in the whole world. Thus, while forecasters are expected to use polite language, overuse of such devices negatively affects how the prediction is perceived.

Participants observed that Inácio used exaggerated flattery and indirect put-downs as a way of simultaneously performing his own humility and authority. For example, there were criticisms of Inácio’s assertion that “nobody knows anything; God is the one who knows.” Inácio has made this statement many times in his predictions at the annual meetings and in general conversations about predictions. While it is ostensibly a demonstration of humility (he often includes himself, saying “I don’t know anything”), this expression is sometimes interpreted as an insult to other forecasters. In particular, the most common explanation given by participants for why FUNCEME no longer attends the annual meeting of the rain prophets was that in 2007, Inácio had said “FUNCEME doesn’t know anything” and the meteorologists, offended, decided not to return in subsequent years.5 The evaluations of Inácio’s prediction demonstrate that people are taking into account not only the accuracy of the forecast but other communicative elements as well.

Problematic aspects of predictions made by other rain prophets were also attributed to the prediction being a public performance. Some people mentioned that having to speak into a microphone and give a prediction in front of a large audience, including the media, caused rain prophets to become nervous and forget what they wanted to say or to have trouble expressing themselves clearly. One rain prophet recounted how he had come to the meeting ready to make a prediction for drought but became so unsure of himself after hearing several optimistic predictions from others that he ended up saying he expected a “good year for rain” though only light, scarce rains. In this case, the desire to be proved right in the future was overridden by the desire to fit his prediction into the immediate interactional context, in which optimistic forecasts were the majority. Ideas about how others will perceive their predictions and identify them—as legitimate rain prophets, liars, God’s faithful, or attention seekers—shape how forecasters perform their predictions in public in ways that are less relevant when speaking privately, when there are more opportunities to influence each listener’s interpretations.

Inácio insisted that 2010 was not a drought year, even in the first week of June when the season was essentially over and there had been almost no rainfall in Quixadá since the interview in mid-May. He claimed that his forecast had been correct, that it had rained enough to get a small harvest for those who worked hard and had not given up. He pointed out that the reservoirs were full and that water delivered by government-funded trucks was not necessary for people who were industrious and stored water in rain barrels. For Inácio, drought is in large part a question of attitude: with persistence, hope, and satisfaction with what one has, he sees a rainy season where others see drought.

4. Conclusions

Examination of these four predictions gives us a better understanding of how predictions are made meaningful in rural Ceará. The analysis shows how judgments about predictions are influenced by the way they are interpreted. It is not surprising that difficulties in understanding some predictions were associated with less favorable evaluations. More importantly, this study demonstrates that, despite forecasters’ attempts to convey a particular message, they are clearly not the sole determiners of what a forecast can or should mean. Interpretive differences can influence perceptions about forecast accuracy, forecaster credibility, and the potential usefulness of the forecast. This holds true for both traditional and scientific forecasts, in that traditional forecasts are not necessarily always easier to comprehend than scientific ones. Rather than a one-way transfer of information from forecaster to user, this study illustrates how meanings are made through the interaction between speaker and listener, both in the initial production and identification of the prediction as a verbalized expectation for the future and in the assessment of its value in light of particular outcomes.

Considering how meaning is jointly produced requires attending to the importance of a prediction’s noninformative elements to those who hear and evaluate it. In the case presented, we saw how forecasters were held responsible for the way they communicate the prediction as much as for the accuracy of the forecast. When problematic predictions were attributed to environmental changes, they were seen as excusable errors, while predictions inappropriately performed drew criticism. Credibility is thus a measure of identity and performance, as well as the quality of the information provided. By taking into account the multiple interpretations and experiences that influence how forecasts are evaluated in local contexts, we gain insights that help us understand why some forecasts are perceived as more or less useful than others. For example, this research shows how predictions perceived to be too specific or overly certain were said to be less credible because they go against the idea that God is the main determiner and causal force for weather and climate conditions. Predictions that acknowledged the inherent unpredictability of seasonal climate (and other natural phenomena) were better received at the time they were first heard, and were evaluated more positively afterward, because the uncertainty was expressed in familiar terms.

This project can be seen as a kind of forecast verification study that examines communicative competence and forecasting skill from the perspective of the user. Such a study is useful because climate forecasts, whether traditional or scientific, are performed in a variety of styles and are all subject to critical evaluation by those who hear them. It is important to recognize that forecasts are both constructed and interpreted as part of larger discourses in which people make their lives meaningful. This broader context in which predictions are made and “used” is just as important to our understanding of how a particular forecast influences decision making as more conventional socioeconomic variables, such as income or education.

For example, when we look at how people talk about predictions in Quixadá, we see that within the discussions of rainfall distribution and forecast accuracy, predictions are interactions in which ideas are formed and challenged about the social identities of individuals and groups—what it means to be an expert, who can be a rain prophet, who the audiences are for predictions—and through which ideas about how the world is or should be are expressed. Through their predictions, rain prophets exemplify the tradition of working the land, maintaining a close relationship with the natural environment and having faith amid struggle. In their explanations for various problematic aspects of the predictions, farmers, forecasters, and business owners express their perceptions of environmental changes, the relationship between humans and nature, the relative powerlessness and fallibility of humans, the value of science and technology, and what is appropriate conduct for public speakers. Predictions are focal points that bring people together to talk about their situation, debate possible actions, offer encouragement, or simply share the burden of worry. In these ways, predictions can strengthen bonds within the community and make it more resilient.

Acknowledgments

The author gratefully acknowledges the contributions of research assistants Carla Renata Braga de Souza and Julio Hércio Magalhães Cordeiro, as well as the anonymous reviewers. Thanks are extended to the Department of Meteorology at FUNCEME, and especially to Namir Melo, for making data available. Assistance with maps was provided by Ed Eastaugh. This research was funded by the University of Western Ontario.

APPENDIX A

Transcripts of Predictions by Rain Prophets

The transcripts given below are taken from a recording of the Annual Meeting of the Rain Prophets, which took place in Quixadá, Ceará, Brazil, on 9 January 2010. Rain prophets have pseudonyms, and other people are identified with initials only. The following conventions for representing naturally occurring spoken language are used throughout:

  • Line breaks indicate natural breath pauses in speech.
  • Commas indicate very short pauses.
  • Question marks [?] are used for questions with rising intonation.
  • Ellipsis […] indicates omitted words.
  • Words in brackets [ ] are added to give further explanation.
  • Parentheses () are used for nonword sounds.

Roberto: Doctor H. the predictions this year

they are various

it indicates little rains

it, a rainy season with a lot of dry spells

there’s rains, they stop and they continue to rain and a short rainy season

according to, I have here some

little signs

they never miss

this common snail from the caatinga [local vegetation]

when the rainy season is normal it gets all

big like this

which then is a sign of lots of rain in January

on the contrary this year, it’s

there are some smaller ones and some bigger ones

it’s a sign of little rains

at the beginning little rain, there will be a dry spell

yeah, that way

it’s not going to be a rainy season with

a lot of rain

that’s what I concluded

the signs of the caatinga

different from our friend L. G.

who has the technological data

who gets it right and is a great scientist

so that’s it

I wanted to also say to our

population here in Quixadá

that L. G. he is such a scientist that he uses

it’s because there wasn’t time

he uses the water at his house

you who are the authorities

6 times

and in his region it’s an abundant region

the coastal region of Camocim

L. don’t give up

you make our meeting here shine

that’s all Doctor H. (applause)

Dr. H.: Pedro and this year?

Pedro: Doctor, first of all

I want to say hello

to the competent authorities of our city

as well as to thank God

for being present with marvelous people

good listeners

who understand and pay attention to what

we say

to all of you who came from away

from Bahia and other states

a good morning from me, my hug and welcome

on behalf of our city

those, who hear us and who listen to us

last year the doctor said, was

governed by the sun

in 2010 it will be governed by the planet

Venus

because

the first full moon of January it will be

together with Aquarius

which is also the planet that brings water

Venus brings what?

Peace, love, health, tranquility, and some

tips such as

love is in the air

love is beautiful and I like it (laughter)

this is the reality

that which does not love does not understand nature well

any one of us who concentrates on

nature will see any other person

on the other side of the world

and I speak with a lot of awareness

because

I ask nature I call and she

shows me

as she showed me this year

the change of the atmosphere

as well as the change of the climatic situation

of the south

mixed with the one from the northeast

of our Ceará

this is the truth

so it indicated an early rainy season

January will rain well

February will rain regular, a little dry spell

and but this is common

now, after the 8th of March

8th of March

that God who is our father creator and

omnipotent

he will order St. Peter to turn on the shower

for all of the entire northeast of Brazil

as it is in the world

no corner will lack water

a very controlled rainy season, for sure

and whoever plants will have a healthy profit

less disastrous and so on

I am certain

that God will not let me lie to a

crowd like this

that is, to the whole world

and for those who know me my name is

Pedro Alfonso da Costa

but known as Pedro in all of Brazil

prophet, singer, radio host, and

joker too

I was analyzing the nest of the joão-de-barro [bird]

which I have at my house

I made my verses

I can say them can’t I doctor?

Audience: You can. (laughter)

Pedro: To close with a key from another

I looked at the nest and spoke like this:

“A mason is the joão-de-barro

who works conscientiously

makes a beautiful house with the door to the west

I say and give and approve

as someone saying to the people

the rainy season is good and powerful

of straw and mud pressed together

he works without tiring

very close to a stream

or even a steep slope

he finishes beating his wings

jumping up on the house

and sings calling the woman of the house”

a hug people and many thanks (applause)

Inácio: I believe that this year

the rainy season has already begun

every year this rainy season of January and

February that’s the way it is

which was the year

that you saw the reservoir fill up in the month of January and February

which was it?

Which was the prophet who saw

now the rainy season is this here and there

the whole month to March

the 3 months will rain

this reservoir is subject

needs to bleed [overflow] and I say why?

Because it was, the most abundance ever was in 2009

because everything, if you mayors

worried about water

the jug overflowed and people with

yokes on their back with jars with everything when God sent that powerful hand

I have 200 signs it’s not only

the corrupião [bird] and the toad, no

it’s 200 signs

that belong to me on the land

I don’t know anything

nobody knows anything

who knows is God pondering

I admire nature and I admire God, I cure

I defend myself from evil

Satan runs away some place else

my path is God (applause)

but I don’t make promises to saints on the wall

I don’t make promises to anyone

I don’t go around flattering anyone

and I’m not crazy either

just a very poor Brazilian citizen

respecting everyone

and today Doctor H.

thank you very much for being polite

bringing attention to Quixadá

which is in the whole world, in Spain

in the whole country

Doctor H., J. S.

thank you very much J. S., you have capability

I respect you Mr. Mayor

and everything Quixadá is the place most respected by the world in the whole world

first it’s the old Cedro and second it’s the rocks and

second it’s the prophets and a respected place

thank you very much people (applause) …

APPENDIX B

Transcript of News Broadcast of FUNCEME Forecast

See appendix A for transcription conventions. This news report was broadcast on 21 January 2010 on the program Jornal Jangadeiro.

Reporter: Despite the approach of the rainy season

Cearenses have been suffering with the hot sun and heat in the capital city

a year ago the scenario was quite different

the rains that fell in Ceará left

thousands of people homeless and displaced

the rainy period in 2009 registered an

average of a 1000 and 3 mm the second largest in the last 30 yr

but according to the meteorologists

the reality this year in Ceará will be much

different from 2009

the prediction is for irregular rainfall for the next 3 months

the estimate is around 45

percent chance

of a rainy season below the annual historical average here in the state

which is 600 and 29 mm.

Meteorologist: The principal factor that we have today

is the El Niño event

which is an event in the Pacific Ocean

in which the water is hotter than normal

this event

it alters the circulation of winds at a global level

making the air descend over Ceará

impeding the formation of rain clouds

REFERENCES

  • Adger, W. N., , Kelly P. M. , , Winkels A. , , Huy L. Q. , , and Locke C. , 2002: Migration, remittances, livelihood trajectories, and social resilience. Ambio, 31, 358366.

    • Search Google Scholar
    • Export Citation
  • Bucholtz, M., , and Hall K. , 2005: Identity and interaction: A sociocultural and linguistic approach. Discourse Stud., 7, 585614.

  • Canal Rural, cited 2010: Previsão do tempo. [Available online at http://www.canalrural.com.br/canalrural/jsp/default.jsp?uf=1%26local=1%26capa=963%26section=previsaodotempo.]

    • Search Google Scholar
    • Export Citation
  • Climatempo, cited 2010: Região nordeste. [Available online at http://www.climatempo.com.br/previsao-do-tempo/regiao/ne/nordeste.]

  • EMATERCE, 2010: Situação da safra agrícola, 31 de Julho de 2010. Ceara State Rural Extension and Technical Assistance Enterprise Rep., 34 pp. [Available online at http://www.ematerce.ce.gov.br/arquivos/SITUACAODASAFRAAGRICOLA31JUL2010.pdf.]

    • Search Google Scholar
    • Export Citation
  • Green, D., , and Raygorodetsky G. , 2010: Indigenous knowledge of a changing climate. Climatic Change, 100, 239242.

  • Henshaw, A., 2006: Winds of change: Weather knowledge amongst the Sikusilarmiut. Climate Change: Linking Traditional and Scientific Knowledge, R. Riewe and J. Oakes, Eds., Aboriginal Issues Press, 177–186.

    • Search Google Scholar
    • Export Citation
  • IBGE, cited 2010: Censo agropecuário 2006: Resultados preliminares. [Available online at http://www.ibge.gov.br/home/estatistica/economia/agropecuaria/censoagro/2006/agropecuario.pdf.]

    • Search Google Scholar
    • Export Citation
  • Ingram, K., , Roncoli C. , , and Kirshen P. , 2002: Opportunities and constraints for farmers of West Africa to use seasonal precipitation forecasts with Burkina Faso as a case study. Agric. Syst., 74, 331349.

    • Search Google Scholar
    • Export Citation
  • INMET, cited 2010: Normais climatológicas do Brasil 1961-1990. [Available online at http://www.inmet.gov.br/html/clima/mapas/?mapa=prec.]

    • Search Google Scholar
    • Export Citation
  • Krupnik, I., , and Jolly D. , Eds., 2002: The Earth is Faster Now: Indigenous Observations of Arctic Environmental Change. ARCUS, 356 pp.

  • Lemos, M. C., , Finan T. , , Fox R. , , Nelson D. , , and Tucker J. , 2002: The use of seasonal climate forecasting in policymaking: Lessons from northeast Brazil. Climatic Change, 55, 479507.

    • Search Google Scholar
    • Export Citation
  • Livingstone, D., 2010: Conversations on rain-making. Perspectives on Africa: A Reader in Culture, History and Representation, R. R. Grinker, S. C. Lubkemann, and C. B. Steiner, Eds., Wiley-Blackwell, 245–248.

    • Search Google Scholar
    • Export Citation
  • McCrea, R., , Dalgleish L. , , and Coventry W. , 2005: Encouraging use of seasonal climate forecasts by farmers. Int. J. Climatol., 25, 11271137.

    • Search Google Scholar
    • Export Citation
  • Meinke, H., , and Stone R. , 2005: Seasonal and inter-annual climate forecasting: The new tool for increasing preparedness to climate variability and change in agricultural planning and operations. Climatic Change, 70, 221253.

    • Search Google Scholar
    • Export Citation
  • Nelson, D., , and Finan T. , 2009: Praying for drought: Persistent vulnerability and the politics of patronage in Ceará, northeast Brazil. Amer. Anthropol., 111, 302316.

    • Search Google Scholar
    • Export Citation
  • Orlove, B., , Broad K. , , and Petty A. , 2004: Factors that influence the use of climate forecasts: Evidence from the 1997/98 El Niño event in Peru. Bull. Amer. Meteor. Soc., 85, 17351743.

    • Search Google Scholar
    • Export Citation
  • Orlove, B., , Roncoli C. , , Kabugo M. , , and Majugu A. , 2010: Indigenous climate knowledge in southern Uganda: The multiple components of a dynamic regional system. Climatic Change, 100, 243265.

    • Search Google Scholar
    • Export Citation
  • Patt, A., , and Gwata C. , 2002: Effective seasonal climate forecast applications: Examining constraints for subsistence farmers in Zimbabwe. Global Environ. Change, 12, 185195.

    • Search Google Scholar
    • Export Citation
  • Patt, A., , Suarez P. , , and Gwata C. , 2005: Effects of seasonal climate forecasts and participatory workshops among subsistence farmers in Zimbabwe. Proc. Natl. Acad. Sci. USA, 102, 12 62312 628.

    • Search Google Scholar
    • Export Citation
  • Pennesi, K., 2006: Aspectos lingüísticos nas previsões de chuva: FUNCEME e os profetas da chuva. Profetas da Chuva, K. P. H. Martins, Ed., Tempo d’Imagem, 144–155.

    • Search Google Scholar
    • Export Citation
  • Pennesi, K., 2007a: Improving forecast communication: Linguistic and cultural considerations. Bull. Amer. Meteor. Soc., 88, 10331044.

    • Search Google Scholar
    • Export Citation
  • Pennesi, K., 2007b: The predicament of prediction: Rain prophets and meteorologists in northeast Brazil. Ph.D. dissertation, University of Arizona, 387 pp.

    • Search Google Scholar
    • Export Citation
  • Roncoli, C., 2006: Ethnographic and participatory approaches to research on farmers’ responses to climate predictions. Climate Res., 33, 8189.

    • Search Google Scholar
    • Export Citation
  • Roncoli, C., , Orlove B. , , Kabugo M. , , and Waiswa M. , 2010: Cultural styles of participation in farmers’ discussions of seasonal climate forecasts in Uganda. Agric. Hum. Values, 28, 123138.

    • Search Google Scholar
    • Export Citation
  • Saville-Troike, M., 2003: The Ethnography of Communication: An Introduction. 3rd ed. Blackwell, 336 pp.

  • Secretaria de Agricultura e Pecuária, 2002: II plano indicativo de desenvolvimento rural do Ceará 1999-2002: Rumo ao desenvolvimento rural do Ceará. Governo do Ceará, 97 pp.

    • Search Google Scholar
    • Export Citation
  • Serviço Geológico do Brasil, 1998: Map of Quixadá. Ministério de Minas e Energia.

  • Stuart, N., and Coauthors, 2006: The future of humans in an increasingly automated forecast process. Bull. Amer. Meteor. Soc., 87, 14971502.

    • Search Google Scholar
    • Export Citation
  • Suarez, P., , and Patt A. , 2004: Cognition, caution and credibility: The risks of climate forecast application. Risk Decis. Policy, 9, 7589.

    • Search Google Scholar
    • Export Citation
  • Wilhite, D., 1993: The enigma of drought. Drought Assessment, Management and Planning: Theory and Case Studies, D. Wilhite, Ed., Kluwer Academic Publishers, 3–15.

    • Search Google Scholar
    • Export Citation
1

Original terms are given in Portuguese, the only local and official language. All research activities were conducted in Portuguese. Translations are by the author.

2

A município is the smallest political and administrative unit, comprising an urban center and the surrounding rural area. Municípios are subdivided into districts to facilitate administration and service delivery.

3

Names given for rain prophets are pseudonyms. Other people mentioned are identified only with initials.

4

See appendix A for transcription conventions.

5

FUNCEME personnel do not support this theory and instead cite the new institutional policy against disseminating forecasts before the official one is produced at the annual workshop, which occurs at the end of January.

Save