Vera Masagao Ribeiro / Antonio A. G. Batista

This article focuses on the challenges related to adult literacy and education. It analyses, first, the evolution of illiteracy rates and of the concepts of literacy used in measurements. Secondly, it identifies the regions and groups that are most affected by illiteracy. It also describes the policies that aim at facing the problem, especially in the field of Adult Education, and the principal tendencies and innovations. Finally, it points out the main challenges for Brazil to achieve the goals set out in its National Education Plan and fulfill its part in the commitments of Education for All, reaffirmed in 2000 at the World Education Forum held at Dakar. Vera Masagao Ribeiro is PHd in Education at the Catholic University of Sao Paulo, Program Director ofAgao Educativa - Brazil. Antonio A. G. Batista is PHd in Education, professor at the Federal University ofMinas Gerais (UFMG); Director of the Center for Literacy Studies (Ceale/UFMG) - Brazil

Commitments and Challenges of a Literate Brazil

In 2001, Brazil's National Congress promulgated, as law, a National Education Plan that defined 26 priority goals to be reached by 2011. Among these are the commitments to offer all children, from the age of 6 on, a nine-year Primary School (at present, it is eight years); to eliminate illiteracy; and to give 50% of those who are 15 years old and over Adult Education corresponding to the first four grades of Primary School in 5 years and up to the 8th grade in 10 years. The goals are ambitious for a country with a population of 170 million and a very large educational debt that has been accumulating for over a century. However, considering that Brazil has a dynamic economy, which is among the 15 largest in the world, the government and Brazilian society are in a position to honour this debt so long as they guide their policies to promote economic development with greater social justice.

Educational rights are amply set out in Brazilian legislation. The Federal Constitution establishes that it is the right of the citizen and obligation of the State to offer 8 years of free Primary Education, including for those who did not have access to it at the appropriate age. Public authorities can be made legally responsible if they do not guarantee this right. The existence of such an advanced legal benchmark does not guarantee, however, that this right is made effective. Certainly, over recent years, education in Brazil has made great progress; the country has managed to almost universalize the access of children to school, achieving a coverage rate of 97.2 % in the age group 7 to 14 in 2003. However, the school system faces great difficulties and has very low productivity. Millions of adolescents drop out of school before finishing Primary School, swelling the contingent of youth and adults with insufficient schooling that number 70.7 million. This was the number of those 15 years old and over that in 2000 had less than eight years of schooling and among these, 16 million were illiterate (IBGE, 2001). The rate of Brazilian illiteracy is greater than that of 70 % of Latin American countries, some of them with a lower per capita income and a lower HDI, as, for example, Peru and Ecuador (INEP, 2003a). Brazil still has, therefore, great challenges in providing literacy and education for all.

The Evolution of Literacy Rates and Concepts

Brazil has had census data on illiteracy ever since the end of the 19th century. Since then the rates have been gradually declining: from 65.3% in 1920, reaching a low of 13.6% in 2000 (Graph 1A). It is not evident in any of the periods that a sharper decline could be attributed to the great campaigns against illiteracy that were carried out throughout the century (Ferraro, 2002). All indications point to the fact that the reduction was a result of the constant expansion of the systems of public education and, secondly, of the gradual gains in the field of Adult Education. It is important to observe also that, due to overall population growth, the absolute number of illiterates only began to decrease during the 1980's (Graph 1B). According to projections made by Henriques (2005) based on data from the 1980's and 1990's, if this pace of reduction is not altered, the rate of illiteracy will be lower than 5% around the year 2005 and in 2020, it will be around 2.5%.

Graph 1A: Evolution of illiteracy rates of the population 15 years and over (Brazil 1920-2000)

Source: Demographic Census 1920-2000 organized by Ferraro, 2002.

Besides the census carried out every ten years, the Brazilian Geographical and Statistics Institute (IBGE) carries out an annual National Homestead Sampling Research (PNAD). In these assessments the rates of illiteracy are calculated based on a self-evaluation made by those interviewed of their capacity to read and write a simple message in the language they know.1 In the 1990's, following a UNESCO recommendation, IBGE began to also publish functional illiteracy rates based on the number of years of schooling. Persons with less than four complete grades are considered functional illiterates. According to this criterion, in 2001 33 million Brazilians 15 years old and over were functional illiterates, which represents 27.3 % of the population in this age group (INEP, 2003b).

Graph 1B: Evolution of illiterate population 15 years and over (Brazil, 1920-2000)

Source: Demographic Census 1920-2000 organized by Ferraro, 2002.

IBGE justifies the indication of the 4th grade as schooling corresponding to functional literacy on the supposition that the rates of regression to illiteracy are very high among people who have not concluded this level of schooling (IBGE, 2001 b). As observed by Wagner (1999), however, the scientific community does not have valid evidence on retention or regression of literacy skills after systematic instruction. Besides this, the concept of functional illiteracy as defined by UNESCO refers to the capacity of facing up to the demands of literacy within the social context (UNESCO, 1978). These demands may vary considerably within the country and it is difficult to make them correspond to a specific school grade. We must therefore consider that the rationale of indicating the 4th grade as the level of functional literacy is more political than scientific; it functions as an educational goal to be achieved.2 A more objective and comprehensive approach to this issue demands methods that permit direct observation of skills and uses of reading and writing, as well as considering a scale of levels or types of literacy, to replace the dichotomy of literate / illiterate.

In Brazil, the first large-scale studies that used skills tests in literacy were executed by evaluations of the school systems. In the 1990's the Ministry of Education organized the National System for Basic Education Evaluation (SAEB), which periodically verifies the performance of the students of all the public and private schools in the country. The system has had data for comparison since 1995 and reveals a disturbing picture with regard to literacy.3 In the evaluation carried out in 2003 (INEP, 2004), only 4.8 % of the 4th grade students achieved adequate performance; the performance of 18.7 % was considered very insufficient (they did not develop reading skills) and of 36.7 % insufficient (they only read simple sentences). In the test given to 8th grade students, which measures more advanced skills, 26.8 % were rated with insufficient performance. This data indicates, therefore, that in spite of the advances in extending schooling, illiteracy continues to reproduce itself in the new generations, within the school itself.

The only indicator of the levels of literacy of the adult population based on skills testing is the initiative of non-governmental organizations: the National Functional Literacy Index (NFLI). With the objective of fomenting a debate and public engagement in the literacy issue, NFLI has been publishing, ever since 2001, the results of annual home surveys with sample groups representing the Brazilian population aged 15 to 64 (Ribeiro, 2003; Fonseca, 2004). Four surveys have been carried out: two for reading and writing and two for mathematics; besides skills testing, detailed questionnaires have been applied on reading, writing and mathematics practices in various contexts: home, work, religious, community participation and continuing education. NFLI uses a comprehensive concept of literacy, understood as

"the skills involved in the use of written language and numeric calculation, and its actual use in social practices, by individuals, social groups and societies, as well as the meaning those individuals and groups attribute to the development of those skills and practices"
(Batista e Ribeiro, 2005, p.3).

According to NFLI testing carried out in 2003 (Instituto Paulo Montenegro & Agao Educativa, 2003), 8 % of Brazilians in the 15 to 64 year age bracket are in a condition of absolute illiteracy and 30% have a rudimentary skill level, that is, they are only able to find simple information in statements of one or two sentences. This group uses reading and writing in a very restricted manner, mainly in the context of religious practices. Another 37% are able to find information in short texts and only 25 % demonstrate full command of the skills tested (locate information, compare texts, use text references, among others). The NFLI carried out in 2004, which focused on mathematics skills (Instituto Paulo Montenegro & Agao Educativa, 2004), verified that the proportion of those that do not manage to read numbers is small, only 3 %, but 29 % only read the numbers that are present within their everyday use (for example, telephone numbers, prices), 46 % are able to solve problems involving a single operation and only 23 % are able to solve problems that involve the linking of more than one operation, such as calculating prices with a discount percentage, besides interpreting simple graphs and maps.

NFLI data suggests that, if it is necessary to adopt a functional literacy criterion based on the level of schooling, the 8th and not the 4th grade of Primary School would be more appropriate, since only among the population that has reached at least the 8th grade, the proportion of those who master the basic literacy skills surpasses 80% (Batista e Ribeiro, 2005). The eight school grades criterion also has a greater political basis since this is the level of schooling that the Constitution guarantees as a right of all citizens. Networks of organizations that do educational rights advocacy in Brazil have guided themselves by this benchmark and have demanded from public authorities the extension of Primary School to the adults who did not have access to it at the appropriate age. Important networks who advocate this are the Youth and Adult Education Forums, and the National Campaign for the Right to Education, which bring together NGOs, teachers' unions, social movements and local education administrators.

The Distribution of Illiteracy

According to the Census 2000, of the total illiterate population, 51.3 % were women and 66.4 % were over 39 years old. Table 1 shows a decrease in illiteracy rates in the younger population, who benefited from the recent extension of schooling opportunities. Table 1 also shows that the advantages of younger groups are higher among women. This is the result of a process that began in the middle of the last century, when the female population began to overcome their educational disadvantage, having succeeded in surpassing the average number of years of schooling of the male population in the 1980's. At present, women occupy the greater part of the enrolments in the regular school system. Many explanatory hypotheses can be raised about this tendency, ranging from the need and possibilities of women to invest more in order to compete in better conditions in a labour market that is unfavourable to them to the greater disposition of women - socially constructed - to adapt themselves to the school culture. In one way or another, this situation has not resulted from any specific educational policy, but from a broader social movement towards the emancipation and equality of women.

Table 1: Illiterates 15 years old and over by age groups and gender (Brazil: 2000)

 Age Groups
 TotalMen Women
N%N %
N
%
15 to 29 years432005 4,03 2870055,32946,002,72
10 to 34 years
13303275,69
837329
7,17
492998
4,22
35 to 39 years1040647
7,52
618652
9,08
421994
6,00
40 to 49 years117912,981
9,19
670639
10,54
5277142
7,91
50 to 59 years1273521,68
10,21
668772
11,23
5834069,25
60 to 69 years 253002213,131211179
1298
1318843
13,26
70 to 59 years2712735
21,68
1195679
19,93
1517056
23,29
60 to 69 years2445134
29,85
1021698
26,98
1423437
32,32
70 years and over 2526433
39,80
1015297
37,05
1511136
12,91
mographi15467262
12,94
7526250
12,96
7941012
12,91

* It does not include those who did not state their age. Source: IBGE, Demographic Census 2000 (IBGE, 2003).

Other factors that determine disparities in the distribution of literacy are intimately related to the extreme inequality of the distribution of wealth that is characteristic of Brazil. In the Northeast region, which is the poorest, the illiteracy rate was 24.9 %, while in the Southeast, the country's main economic hub, it was 7.2% in 2000 (IBGE, 2003).4 In the urban areas, the rate was 9.5 % as opposed to 28.3 % in the rural areas. Due to population concentration, lower rates can represent larger groups; Sao Paulo, for example, with a rate below 5% was the city with the greatest number of illiterates - 383 thousand (INEP, 2003b). At any rate, one cannot lose sight of the fact that it is more difficult to combat illiteracy in regions that have fewer resources. Graphs 2A and 2B give an overview of the regional distribution of illiteracy in the country and show how different the situation can seem when rates or absolute numbers are considered.

Graph 2A: Illiteracy rates of the population 15 years old and over Brazil (2000)

* It does not include those who did not state their age. Source: IBGE, Demographic Census 2000 (IBGE, 2003).

Another disparity in the condition of literacy, strongly related to the disparities in the access to wealth, refers to the ethno-racial groups that make up the Brazilian population, of which 53.6% are whites, 45.3% are Afro-descendents (blacks and mulattos), 0.6% oriental, and 0.5% indigenous. Within the illiterate population, the Afro-descendants have a proportional ly larger partici pation (62.9 %), as do the indigenous (0.8 %), while whites and Orientals have a smaller participation, 35.3 % and 0.2 % respectively (IBGE, 2003). The illiteracy rate among indigenous people is the biggest (30.2 %), followed by blacks (23.2%), mulattos (21.1 %), whites (10.9%) and Orientals (6.6%). Considering the relations between Afro-descendents and whites, the two majority groups, inequalities related to literacy can be observed in learning assessments and in average years of schooling. According to SAEB's data, the proficiency of Afro-descendent children in literacy tests is lower than that of white children even when their parents have the same levels of education (Araujo & Luzio, 2003). According to a study carried out by Henriques (2001), although both groups have had a gradual increase in the average years of schooling during the 20th century, comparison of different generations shows the persistency of inequality: young Afro-descendents born in 1974 have an average 2.3 years of schooling less than young whites born in the same year, and this difference is the same observed in the schooling average of their parents and grand-parents. Based on these data, the author points out the inertia of discrimination patterns in our society.

Graph 2B: Illiterates 15 years old and over Brazil (2000)

* It does not include those who did not inform their age. Source: IBGE, Demographic Census 2000 (IBGE, 2003).

Over recent years the country has undertaken some initiatives in order to accelerate combating ethno-racial inequality in education. Some public universities guarantee quotas of vacancies for Afro-descendents, which might have a motivating effect on that group of students in basic education. A law approved by the National Congress in 2004 makes obligatory the inclusion of contents of Afro-Brazilian history and culture in basic education curricula, which can also favor a feeling of belonging of that group in relation to school culture.

Literacy and Adult Education

The Brazilian educational system is highly decentralized. The availability of basic education is mainly the responsibility of the 26 States and over 5 thousand municipalities that make up the federation. The responsibility of the federal government is that of coordinating, evaluating and promoting equity. Private schools also participate in the system, but around 90 % of students in basic education study in public schools that are free and where no kind of fees can be charged.

The legal organization of Brazilian education recognizes adult education as a form of specific schooling, to which is attributed greater flexibility as to work load, curriculum, forms of evaluation and use of distance learning methodologies. At the Primary School level this is geared to the population 14 years and over. This form of schooling, presently called Youth and Adult Education (YAE), is provided mainly by municipal and state school systems, and the same professionals that teach in regular education attend the students. Besides this, some state and municipal governments as well as the federal government and civil society organizations also maintain adult literacy programs outside of the school systems, employing non-professional monitors. The YAE enrollments inside the school systems are accounted for in the School Census carried out annually by the Ministry of Education, which gathers information provided by registered schools all over the country. Nevertheless, the School Census does not include literacy programs outside the school systems so the best source of data related to participation on this kind of program is the PNAD, the official National Household Survey. PNAD tallies information provided by heads of households about household members' frequency at every level of schooling and in adult literacy programs during the period in which the survey is being carried out.

According to the PNAD, the number of people attending adult literacy programs has increased over the last decade, especially from 1996 and then in 2003, when the federal government took relevant initiatives in this area (Graph 3). In 2003, there were 786 thousand people attending adult literacy courses, which means 5.3 % of the number of people that declared they were illiterate in the same year (14.6 million); 72 % of the participants lived in urban areas and 59 % in the Northeast region, 63 % were women, 71 % Afro-descendents and 71 % were 35 years old and over (SECAD / MEC 2005).

Graph 3: People frequenting Adult Literacy courses (Brazil 1992-2003)

Source: PNAD 1992 - 2003, organized by SECAD I MEC 2005

Enrollments in the YAE equivalent to Primary School also increased over recent years, from 2.0 million in 1998 to 3.4 million in 2004, according to the School Census. During this period a strong tendency was the municipalizing of these enrollments: in 1998, the municipal school systems were responsible for 30.2 % of the total and in 2004, for 58.1 % (Table 2). Decentralization is a general tendency in all of Primary School, resulting in policies of educational reform that have dominated since the 1990's (Di Pierro, 2001).

Table 2: Evolution of initial enrollments in Youth and Adult Education - Primary Education (Brazil, 1995-2004)

YearTotalFederal
StateMunicipalPrivate
N
%
N
%
N
%
N
%
19982081710561
0.2
1316533
63.3
.09865030.2
134957
3.5
200126368884885
0,2
123898947,0
126774048.1
125274
4.8
2002
2788113
2733
0.1
1098828354
19877235
57.0
98650
3,5
2003
3315887
909
0.0
1387505
41.8
1846964
55.7
80509
2,4
2004
3419170
381
0.1
1354303
39.6
1987723
58.1
76763
2.2

Source: School Census 1998-2004 (INEP, 1999, 2003 and 2005)

Even though legislation allows for a type of specific education for youth and adults, there are a considerable number of students 15 years and over enrolled in Regular Education. In theory, Regular Primary Education should cater for students in the regular school age bracket (7 to 14 years old). Besides the problem of the youngsters that take more time than expected to finish Primary School, in recent years more and more enrollments of older people have been registered in Regular Education. This distortion is due to the mechanisms that rule the financing of education in the country, which does not stimulate States and Municipalities to offer YAE, as will be explained further on. Considering the existence of this distortion, in order to estimate the coverage of youth and adult educational demands, it is necessary to consider enrollments in YAE and in Regular Education. Table 3 compares the number of enrollments in YAE and of students older than 14 years in Regular Education with the number of youngsters and adults that had less than eight grades of schooling in 2000. This comparison can give an idea of how distant are educational opportunities from the demands of youth and adult population and of our National Education Plan goals.

Table 3: Population 15 years old and over by level of schooling and enrollments in Youth and Adult Education and Regular Education (Brazil, 2000)

Population 15 years old and over (A)
Enrollments of students 15 years old and over
B+ C/A

%


Youth and Adult Education (B)
Regular Education (C)
Level of schooling
Age groups
n.
1st to 4th grade
5th to 8th grade
1st to 4th grade
5th to 8th grade
No
15 to 245347268
394091 -1419148
-
33.9
schooling to 3ri grade
25 and over
27873991
403210
-
330902
-
2.6
From 4th to 7th graphicCe
15 to 24
11621139-
821276
-
6176694
60.2
25 and or
25949005
-
572084
-
457257
4.0
TOTAL
 70791403
797301
1393360
1750050
6633951
14.9

Sources: For (A), Demographic Census 2000 (IBGE, 2003) and for (B) and (C) School Census 2000 (INEP, 2005)

Literacy Programs: The Partnership Model

Throughout the 20th century, as the regular basic education systems began to be organized and expanded, adult education began to be the focus of initiatives that took on the form of campaigns, offering short-term courses and mobilizing non-professional literacy monitors. Among the more significant campaigns of national scale are those of 1947-1950, which catered for 830 thousand persons in 1950 (Beisiegel, 1974), and the Brazilian Literacy Movement - Mobral, which covered 7.3 million between 1970 and 1972 (Correa, 1973). Although these campaigns did not have significant effects on the acceleration of the rates of literacy in the country, they certainly contributed towards the gradual organization of the system of YAE, both by stimulating the demand for education in this segment of the population and by involving local authorities in providing this kind of education. Even though it is far below the demand, the YAE system has been making it possible for many adults to raise their schooling level.

At present the efficacy of short-term literacy courses is under suspicion, especially when the learners come from social environments with low levels of literacy. Even in Regular Education it is assumed that literacy is a long term process, and that is why in many school systems allowance has already been made for cycles of two to three years for the children to achieve basic skills related to literacy.5 The majority of adult educators also testify that short-term programs do not guarantee a minimum of learning skills. Statistics offer strong indicators for this: the PNADs carried out from 1992 to 2002 show that, among people who declare that the highest level of education that they have reached was an adult literacy program, about 60 % of them declare that they are illiterate (Henriques, 2005). The NFLI has also made evident that among adults with 1 to 3 years of schooling, the percentage of those who do not even demonstrate rudimentary skills is 32%.

In spite of these evidences, the initiatives characterized as literacy campaigns have not lost their political appeal. Probably, governments still consider that short term initiatives against illiteracy are a better strategy to advertise their good intentions, to mobilize society or to cover more people in a short period, while providing basic education for all would cost substantially more and would have more gradual and long term results. Nonetheless, at least two novelties in the literacy program policies can be identified in the 1990's: the adoption of strategies based on the involvement of local governmental and nongovernmental partners and the recognition, at least formally, that the effort to teach youth and adults to read and write must be followed by offering opportunities for this public to continue their education.

At the beginning of the 1990's an initiative in the city of Sao Paulo had a national impact. Paulo Freire, the most internationally renowned Brazilian educator, had taken office as Secretary of Education of that city. Updating the literacy principles and methodology that had made him famous, Paulo Freire launched the Literacy Movement (MOVA), establishing partnerships with community based organizations. These organizations were responsible for mobilizing learners and their monitors, as well as providing rooms where the groups could be taught. The city government would provide a stipend for the facilitators and supervise the development of the groups that met for two hours, four times a week. The monitors met weekly with pedagogical supervisors to plan their work. There was no specific time established for the students to remain in the groups and for those who felt prepared, a vacancy in the YAE municipal system was guaranteed so that they could continue their studies. In 1992 the program covered over 18 thousand youngsters and adults in partnership with 73 community organizations (Pontual, 1995). Due to changes in the municipal government, the program was interrupted at this time and was taken up again only in 2001. In 2004 it was catering for 32 thousand youngsters and adults in partnership with 160 civil society organizations.

The MOVA model was disseminated throughout other cities, assuming specific characteristics and involving new partners, such as unions and companies, but maintaining "Freireian" inspiration as a common link (MOVA, 2004). In 2003, a national network that promotes annual meetings - MOVA Brazil - gathered managers from five States and approximately 20 cities from all the regions of the country (Di Pierro e Graciano, 2003). Many MOVAs produced materials recording the experiences that were developed but, unfortunately, no systematized information was produced referring to the learning results and impact on learners living conditions or on how many of them continued their studies in YAE.

The federal government only resumed taking on an important initiative in adult literacy in 1996, when it launched the Alfabetizagao Solidaria Program (PAS), drawn up by a body specially created to coordinate policies to fight poverty. The strategy adopted by the PAS was also that of partnerships, encouraging the participation of municipal governments, companies and Higher Education Institutions (HEI). Initially, it was geared only to the poorer regions of the North and Northeast, but later it also began to operate in large urban centers of other regions. Different from the MOVA model, the PAS is organized in six-month modules, the first of which is dedicated to planning and training monitors and the others to the literacy activities themselves that total a 240-hour workload. As a principle, each learner and each facilitator can only participate in one module while in the ensuing semesters new groups must be established and new monitors trained. The HEIs coordinate the work of mobilizing the municipalities and also train and give pedagogical supervision to the monitors, selecting them from the local population. The municipal authorities are responsible for convoking participants and indicating the place where the groups can work.

As of 1998 PAS was legally constituted as a non-governmental organization, Alfabetizagao Solidaria, but it continued receiving the greater part of its resources from the federal government. It also began to sign agreements with State governments and to organize campaigns to appeal to individual donors, using publicity through the media, with the support of artists and celebrities. Until 2002 this non-governmental organization channelled a great amount of federal funds for YAE. According to the information published by the organization, from 1997 to 2004,4.9 million students were covered, in 2050 municipalities, with the partnership of 144 companies and 209 HEIs. The cost per student was R$ 168.00 (US$ 61.70),6 resources used to give grants to the facilitators and local coordinators, lunches and schoolbooks for the students, training and evaluation (Alfabetizagao Solidaria, 2004).

In 2003 the new management that took office in the federal government decided, at the very beginning of their mandate, to recover the leadership of the initiatives in this area and launched the Literate Brazil Program (PBA) with the goal of eliminating illiteracy in four years, that is, at a more accelerated pace than that anticipated in the National Education Plan. With the use of agreements, the PBA gives financial support to governmental and non-governmental agencies that already have experience in adult literacy so that they can expand their coverage. In 2003, with a investment of R$ 175.5 million (US$ 59.9 million) the federal government transferred funds to the signees on the basis of R$ 15.00 (US$ 5.12) per month per learner to cover paying the monitor for six months, plus another R$ 80.00 (US$ 27.32) to train each one. The organizations that signed agreements were held responsible for investing the remaining resources that would be necessary, according to their criteria and possibilities. In that year 1.96 million learners were covered.

In 2004, the Ministry of Education was restructured and the Literate Brazil Program (PBA) was transferred to a new Secretariat specially set up to promote continuing education, literacy and diversity (SECAD). With the broader objective of combating exclusion and inequality, SECAD began coordinating literacy and adult education activities, integrating these with programs of rural education, environmental education, indigenous people's education and ethno-racial diversity.

Some modifications were introduced in the PBA during that year: the maximum period for literacy was extended from six to eight months and, so as not to discourage the formation of small groups, necessary to meet the demands of rural and slum areas, a new criterion was used to calculate the monitors' pay: a base pay of R$ 120.00 (US$ 44.07) plus R$ 7.00 (US$ 2.57) for each student per month. The cost for the training of each facilitator, raised to R$ 120.00 (US$ 44.07), plus the cost of these changes, aimed at optimizing results, resulted in a slight fall in the number of students covered in 2004 - 1.81 million, with a total of R$ 168 million (US$ 61.7 million) invested by the federal government in the agreements. There is no data available to estimate the magnitude of the investment made by the other partners of the PBA, but taking only the investment of the federal government, the maximum cost per student, on the basis of an average class of 20, was R$ 94,00 (US$ 32.10) in 2003 and R$ 110,00 (US$ 40.40) in 2004,

In 2004, the PBA also expanded the participation of State and municipal governments in the partnerships, and the participation of NGOs was no longer in the majority (Graph 4). It is important to note that in 2004 55.9 % of the provision by the non-governmental sector was made by only three NGOs: Alfabetizagao Solidaria, SESI - Industry Social Service - and the Brazilian affiliate of Alfalit- an international Christian organization. For 2005 the PBA management plans to decentralize the program, extending the participation and autonomy of sub national government.

Graph 4: Distribution of students covered by the Literate Brazil Program, according to the type of signee organization (2003-2004)

Source: Henriques (2005)

The establishment of SECAD has favored integration of literacy and YAE policies. With the objective of meeting the demand for continuing studies generated by PBA, in 2004 SECAD invested R$ 420 million (US$ 154.41 million) through agreements with States and municipalities so as to strengthen and expand their YAE networks.

The examples of MOVA, Alfabetizagao Solidaria and the PBA make evident that the partnership model has become the preferred strategy to face the problem of illiteracy in a country with the dimensions of Brazil. On the other hand, it also demonstrates that, within the broad partnership concept, policies with different interpretations of the public nature of these policies and the problem of central versus local can be concealed. In the first place, since illiteracy reaches to a greater degree the less favored social sectors and the poorer regions, it is fundamental that public authority and its central administration invest so as to be able to correct the lack of resources in the poorer regions. Sub national or local public authorities must, in their turn, have more capacity to implement actions that are considered peculiar to each context. In many cases local governments need technical assistance, but it is important that they participate in the management of the literacy program and structure themselves to offer the continuation of studies. The participation of civil society is undoubtedly welcome, both because it complements state efforts and because it is an incentive to participation and to civil society control over public policies. The Brazilian examples that have been described demonstrate that in the large national programs, the tendency is to privilege the larger organizations, connected to companies and churches. If literacy is conceived as a way to empower the less favored social sectors, the participation and empowerment of the community-based organizations must also be targeted. The comparison between the MOVA model and the PBA model suggests that the opportunities of community-based organizations are favored when local governments mediate this.

A recent research carried out in Brazil confirms that, since the 1990's there has been significant growth in the number of registered NGOs, especially those aimed at development and the defense of rights (IBGE, 2004). This is a very valuable social resource that cannot be wasted. However, the effective empowerment of civil society organizations depends on the capacity of the governments to guarantee accountability as to the criteria and results of partnerships that involve the transfer of public funds. Governments cannot simply transfer resources, renouncing their responsibilities for conducting and evaluating public policies.

Evaluation: The Main Challenge for Perfecting Policies and Practices

In the 1990's studies on the psychogenesis of writing (Ferreiro, 1986) as well as researches on literacy as seen from a social perspective (Street, 1984; Soares 1998) gave cause for criticism of school literacy practices based only on methods of identification and association of graphemes and phonemes. Proposals were disseminated that encouraged the diagnosis and extension of students' previous knowledge and hypotheses on the system of writing. Value also began to be given to the use of texts of several genres that could offer living experiences and reflection on the social uses of writing. Advice of this kind was added to some principles of popular education of Freirean inspiration which, since the 1960's, had proposed literacy as a means for adults to become aware of social problems and engage in overcoming them. Parallel to this, literacy proposals based on traditional methods continued to be renewed.

At present, some programs adopt a specific methodological proposal and others do not; in some cases, schoolbooks are used and in others they are substituted by material produced by the facilitators themselves. However, although principles and methodological guides are stated in the literacy programs, it is very difficult to assess whether or not they are carried out in practice. The heated debates as to which is the best method, about how much time is needed for initial literacy to take place or about its individual and social impact, lack empirical evidence with sufficient broad validity on which to base consensus and decision-making. Academic investigations on this theme are normally restricted to case studies conducted by agents who are directly involved with the initiatives (Haddad, 2002), the results of which are usually difficult to generalize even though they do furnish elements for conducting extensive research on a larger scale. The practice of evaluating programs is very new in Brazil and this is one of the main obstacles that must be overcome in order to perfect literacy policies and practices.

Recalling the experience of the literacy programs described in the previous section, it is possible to sketch out the steps that have been taken towards this in the country. With regard to MOVA, it is necessary to recognize that the lack of concern in producing and publishing evaluation results is a weakness which, if overcome, could strengthen these experiences as a model of public policy for literacy.

Alfabetizagao Solidaria carried out some advances in this area by proposing, since its founding, an evaluation system that registers, at each module, the initial and final stage of learning for each student, based on criteria such as knowledge of the alphabet, and capacity to read or write words, sentences and simple texts. As of 2000, they also began to distinguish skills related to orality and to mathematics. The information is given by the facilitators, collected and analyzed by the HEIs and by the program coordination that publishes periodical reports. These documents show that when the learners begin, they have the command of various levels of those capacities; they then progress throughout the course, but only a minority, around 18 % to 27 %, were able to read and write small texts at the end of the courses carried out in 1998 and 2000 {Programs Alfabetizagao Solidaria, 1999; Klein, 2000 and Janeiro, 2002). Faced with these results Alfabetizagao Solidaria argues that its main objective is to mobilize youth and adults to learn and to encourage the municipalities to offer continuing education programs. Referring to complementary studies, Alfabetizagao Solidaria has pointed out, as indicators of program results, the decrease in illiteracy indicators in the municipalities that have been covered, an increase in the offer of YAE, as well as greater participation in community activities (Janeiro, 2002). Independent evaluation carried out by the Federal Fiscal Court, however, was not as optimistic in judging the capacity that the program had to promote continuity of studies and suggested that the literacy period should be extended to improve the employment of the trained monitors and the use of the didactic material as well as to diminish the problem of repeated enrollments (Tribunal de Contas da Uniao, 2002). Even though not formally permitted, many learners are enrolled again, which distorts data concerning the number of beneficiaries in the program.

The PBA, ever since its launching, has registered all the participants as a regular procedure. Personal data of the learners are collected by the signee organizations and gathered by the federal government in a computerized platform. This procedure permits control of cheating in enrollments besides affording an initial basis for future follow up while allowing for interlinking with other social programs. Throughout 2004 PBA developed a broad evaluation proposal that considers several political aspects. It foresees data collection that would permit them to analyse in what measure the investment in the program complements, stimulates or substitutes the investment in other levels of government and of non-governmental organizations. With this objective a national mapping of all the initiatives geared to adult literacy in the country was begun. The evaluation plan intends to use methodologies that are scientifically valid and that permit the identification of long-term results and impacts, isolating the contribution of the program in the changes that are observed. It also intends to collect evidence that validates decisions with regard to the best political and pedagogical strategies to optimize results. Toward this end standardized instruments are being elaborated for cognitive evaluation of sample groups of learners at the beginning and the end of the process, besides the use of questionnaires to be applied during the program and at later dates. Information will also be collected from groups of non-participants and it is expected that it will be possible to constitute random groups for treatment and control.

The plan is ambitious but, even if only partially carried out, it will represent a significant advance. It is a pity that the first steps towards its implementation were only taken at the beginning of 2005, two years after the program was launched, after over 3 million people had already been covered and more than R$ 300 million invested. Maybe more pragmatic proposals, which would offer more immediate information and which could be gradually complemented, would respond better to the urgent need to create an evaluation culture for public policies in Brazil.

Financing Literacy for All

Public expenditure on education in Brazil corresponded, at the end of the 1990s, to 4.3 % of the GDP, a proportion similar to countries such as Argentina, Chile and Paraguay. However, the value of the Brazilian GDP per capita is much smaller and the expenditure per student in Primary School in Brazil, in 1999, represented less that half of that effected by Argentina and Chile (INEP, 2003a). On approving the National Education Plan in 2001, the legislative authority included an article that foresees the progressive expansion of public expenditure on education, reaching 7 % of the GDP by 2011. Unfortunately the executive authority vetoed this article and the Plan that has been enforced at present does not foresee sources of funds that would guarantee the fulfillment of its goals, which has compromised its political consistency.

Adult Literacy and YAE are particularly affected by insufficient funding. International aid in this area is practically irrelevant in Brazil. Data from 1995 indicates that YAE corresponded to 1.4 % of public expenditure on education, considering the three spheres of government (Abrahao e Fernandez, 1999). According to the same source, the federal government was, in that year, responsible for 9.2 % of expenditure on YAE, while the states and municipalities contributed 62.3% and 28.4%, respectively. For more recent periods there is no data that would permit us to estimate the totality of public funds invested in YAE in Brazil, a problem that adds to an even greater insufficiency of information regarding the non-governmental investment in this sector.

It is a fact, however, that as of 1996, Brazil adopted a system for distributing public funds for Primary Education that limited the states' and municipalities' investment in YAE, the main providers of this kind of education. This is the Fund for Development of Elementary Education and Teacher Valorization (FUNDEF), a mechanism that proposes to protect the investment in Primary Education and diminish the disparities of the educational services offered by poorer and richer states and municipalities. The states and municipalities are obliged to collect the greater part of their resources earmarked for education in state funds that are later redistributed according to the number of enrollments that each one has in Primary Education. The role of the federal government is to supplement the funds of the states that do not reach a minimum value of spending on each student as established in law. In the formulation of FUNDEF that was originally approved by the National Congress, YAE could be benefited by the resources from this fund. However, a veto from the executive authority ended up excluding YAE from the FUNDEF. This exclusion dissuaded the states and municipalities from investing in adult education and, in part, explains why in the poorer states that receive FUNDEF supplementary funds, there is such a large number of students over 14 enrolled in the regular education system.

In an attempt to compensate this situation the Federal Government has been maintaining, since 2001, a program for transferring funds geared to the 14 poorest states of the North and Northeast and to 398 municipalities with a low HDI, which receive amounts proportionate to the number of enrollments that they maintain in YAE. This program has involved in an increase of more than 500 % in the federal budget assigned to this type of education. This program has been encouraging the expansion of the offer of YAE in these regions, but this is as yet very insufficient to respond to the demand.

Since 2003 the Ministry of Education has been negotiating with state and municipal governments the setting up of a fund (the Fund for Development of Basic Education - FUNDEB) that would include all levels of basic education and YAE. The National Campaign for the Right to Education enjoins that minimum values that are spent per student should correspond to the real needs of each level and type of education. For this it is fundamental that the sum of public funds geared to education be increased.

The impediments that the executive authority has imposed on the legislative efforts to expand public investment in education correspond to economic policies adopted by recent federal governments. Following the recommendation of the International Monetary Fund, the country has been generating annually a surplus that is superior to 4 % of the GDP, to pay off the internal and external public debt (a percentage close to that which is expended annually on financing all of public education in the country). Unless the binding effect of this economic policy be seriously scrutinized it will not be possible to achieve educational goals for regular education and, even less so, those related to youth and adult education and literacy. The National Campaign for the Right to Education enjoins the National Congress to overthrow the presidential veto of the Article in the National Education Plan that foresees the increase of public expenditure in education to 7% of the GDP in 2011.

Conclusion

Over recent decades Brazil has managed to make important headway in making its population literate. Practically all children have access to schooling, and educational opportunities for youth and adults are also increasing, although at a slower pace. Decentralization as well as local government and civil society partnerships give a new dimension to the literacy programs that can thus answer more adequately to the interests and needs of the communities. The fact that these programs acknowledge that literacy is a long-term process associated to schooling could stimulate a more accelerated expansion of the offer of Youth and Adult Education.

Although one can observe headway, there are challenges that persist. Social and regional inequalities in the country continue to express themselves in an unequal access to literacy: the poorer social groups and regions concentrate the highest illiteracy indicators. It is alarming that illiteracy still predominates among the Afro-descendent population to whom the country has a historic debt. If the problems of YAE financing and the deficiencies of the regular school remain unchanged the country will probably not achieve the goals of its National Education Plan by 2011. Finally, the interlinking of educational policies with policies that aim at combating poverty and a more even distribution of wealth is the fundamental condition for the population as a whole to effectively take advantage of the possibilities that literacy offers for human and social development in all spheres of living.

 

Notes

1 In the first decades of the 20th century, the Census considered literate a person who declared the ability to read and write; in 1950, the IBGE adopted a more precise definition, considering literate a person who knew how to read and write a simple message in a language known by him/her and excluding those who could only sign their names. With little variations, this definition has remained since then (Ferraro, 2002). The Census 2000 report gives the following explanation "Is considered literate a person who can read and write a simple message in a known language." Those who have learned to read and write but have forgotten and those who could only sign their names were considered illiterates (IBGE, 2003, p. 32). According to Ferraro (2002), the regularity of the decline in illiteracy rates during the 20lh century suggests that these changes in definition did not affect strongly the comparability of data.

2 That is why developed countries use the 8lh or 9lh grades as the threshold for functional literacy. Soares (1992), Hillerich (1978) and Newman & Beverstock (1990) criticize the arbitrariness of defining literacy based on a certain number of school grades.

3 SAEB carries out evaluations on "Mathematics" and "Portuguese". Nonetheless, the last evaluation focuses on reading skills (grammar and orality are not evaluated). That is why different authors (even the researchers who carry out the evaluation) take the results of SAEB's "Portuguese" test as an indicator of different levels of literacy skills. See, for example, Bonamino, Coscarelli e Franco (2002), and INEP (2003a).

4 The Human Development Index (HDI) of Brazilian states, published by United Nations Development Programme and IPEA (PNUD & IPEA, 2003), can give an idea of regional disparities in Brazil; The HDI-M of the seven states of the North region vary from 0.697 (Acre) to 0.753 (Amapa); among the nine states of Northeast region from 0.636 (Maranhao) to 0.705 (Pernam-buco and Rio Grande do Norte); among the four states of the Southeast region from 0.765 (Espirito Santo) to 0.820 (Sao Paulo); among the three states of the South region from 0.787 (Parana) to 0.822 (Santa Catarina); and among the three states and the Federal Disitrict of the Central-West region from 0.773 (Mato Grosso) to 0.844 (Federal District).

5 The history of the transformation of the concepts of literacy in Brazil is analyzed by Soares (1998).

6 Conversions to US dollar were calculated considering average exchange rates for the year.

References

ABRAHAO, Jorge & FERNANDES, Maria Alice (1999). Sistema de Informagoes sobre os
gastos publicos da area de Educagao - SIGPE: Diagnostico para 1995. Texto para
discussao n. 674. Brasilia: IPEA - Instituto de Pesquisa Economica Aplicada.

ALFABETIZAQAO SOLIDARIA (2004). Trajetoria 2004 - 8 anos: Responsabilidade e
inclusao social. Brasilia: Alfabetizagao Solidaria.

ARAUJO, Carlos Henrique & LUZIO, Nildo (2003). Saeb e equidade. Boletim do Salto
para o Futuro. Saeb: inclusaopela qualidade. Rio de Janeiro: Secretaria de Educagao
a Distancia do Mec/TVE.

BATISTA, Antonio A.G. e RIBEIRO, Vera M. (2005). Cultura escrita no Brasil: modos
e condigoes de insergao. Educagao e Realidade - numero especial "letramento".
(no prelo).

BEISIEGEL, Celso (1974). Estado e educagao popular: um estudo sobre a educagao
de adultos. Sao Paulo: Pioneira.

BONAMINO, A., COSCARELLI, C. e FRANCO, C. (2002) Avaliagao e letramento: concepgoes de aluno letrado subjacentes ao SAEB e ao PISA. Educagao e Sociedade,
v.23, n. 81, p 91-114, dez.

CORREA, Arlindo (1973). Mobral: Realizagoes e perspectivas. Revista Educagao, n.
2, vol. 7.

Dl PIERRO (2003). Analfabetismo e alfabetizagao: desafios do Programa Brasil Alfabetizado. São Paulo: Ação Educativa.

Dl PIERRO, Maria Clara & GRACIANO, Mariangela (2003). A educagao de jovens e
adultos no Brasil: Informe apresentado a Oficina Regional da UNESCO para America
Latina y Caribe. Sao Paulo: Agao Educativa.

Dl PIERRO, Maria Clara (2001). Descentralizagao, focalizagao e parceria: uma analise
das tendencias nas polfticas publicas de educagao de jovens e adultos. Educagao
e Pesquisa, Sao Paulo, v. 27, n. 2, p.321-337, jul./dez.

FERRARO, Alceu (2002). Alfabetismo e nfveis de letramento no Brasil: o que dizem os
Censos? Educagao e Sociedade, v.23, n. 81, p 21-47, dez.

FERREIRO, Emilia & TEREROSKY, Ana (1986). Psicogenese da lingua escrita. Porto
Alegre: Artes Medicas.

FONSECA, Maria Conceigao Ferreira (org.) (2004). Letramento no Brasil: habilidades
matematicas. Sao Paulo: Global.

HADDAD, Sergio et al. (2002). Educagao de Jovens e Adultos no Brasil (1986-1998).
Serie Estado do Conhecimento no 8. Brasilia: MEC/INEP/COMPED.

HILLERICH, R.L (1978) Toward an Assessable Definition of Literacy. In: CHAPMAN,
L.J., CZERNIEWSKA, P. (Eds.). Reading: from Process to Practice. London and
Henley: Routledge & Kegan Paul in association with the Open University.

HENRIQUES, Ricardo (2001). Desigualdade racial no Brasil: Evolugao das condigoes
de vida na decada de 90. Textos para discussao n. 807. Rio de Janeiro: IPEA - Ins-
tituo de Pesquisa Economica Aplicada.

HENRIQUES, Ricardo (2005). Oficina 1: Avaliagao do Programa Brasil Alfabetizado.
Brasilia: Secretaria de Educagao Continuada, Alfabetizagao e Diversidade / MEC.

IBGE (2001). Sfntese dos indicadores sociais 2000. Rio de Janeiro: Instituto Brasileiro
de Geografia e Estatfstica.

IBGE (2003). Censo Demografico 2000; Educagao: Resultados da Amostra. Rio de
Janeiro: Instituto Brasileiro de Geografia e Estatfstica.

IBGE (2004). As fundagoes privadas e associagoes sem fins lucrativos no Brasil2002.
Rio de Janeiro: Instituto Brasileiro de Geografia e Estatfstica.

INEP (1999). Sinopse estatfstica da educagao basica: Censo escolar 1998. Brasilia:
Instituto Nacional de Estudos e Pesquisas Educacionais Anfsio Teixeira/ MEC.

INEP (2003a). The challengeof quality education for all: Education in Brazil -1990-2000.
Brasilia: Instituto Nacional de Estudos e Pesquisas Educacionais Anfsio Teixeira /MEC

INEP (2003b). Mapa do analfabetismo no Brasil. Brasflia: Instituto Nacional de Estudos
e Pesquisas Educacionais Anfsio Teixeira/ MEC.

INEP (2003c). Sinopse estatfstica da educagao basica: Censo escolar 2002. Brasflia:
Instituto Nacional de Estudos e Pesquisas Educacionais Anfsio Teixeira/ MEC.

INEP (2004). Resultados do SAEB 2003 Brasil: Versao Preliminar. Brasflia: Instituto
Nacional de Estudos e Pesquisas Educacionais Anfsio Teixeira / MEC.

INEP(2005) Sinopse estatfstica da educagao basica: 2000-2001-2003 e Censo escolar
2004. www.inep.gov.br, Jan, 14.

INSTITUTO Paulo Montenegro, Ação Educativa (2004). Indicador Nacional de Alfabe-
tismo Funcional - um diagnostico para a inclusao social pela educagao. Instituto
Paulo Montenegro / Agao Educativa: Sao Paulo.

INSTITUTO Paulo Montenegro, Agao Educativa (2004). Indicador Nacional de Alfabe-
tismo Funcional - conhecimentos matematicos. Instituto Paulo Montenegro / Agao
Educativa: Sao Paulo.

JANEIRO, Cassia (2002). Programa Alfabetizagao Solidaria : aumento de matrfculas
na Educagao de Jovens e Adultos - Censo Escolar 2000 - 2001, resultados comprovados pelo IBGE, impacto nos municfpios e nas instituigoes parceiras, avaliagao do modulo VIM. Brasflia PAS.

KLEIN, Lygia et al (2000). Programa Alfabetizagao Solidaria : perfil dos alunos, impacto
em 23 municfpios do Projeto Piloto, avaliagao do modulo VII. Brasflia, PAS,.

MOVA Brasil (2004). MOVA em Movimento. Sao Paulo: Instituto Paulo Freire, Agao
Educativa.

NEWMAN, A.P., BEVERSTOCK, C. (1990). Adult Literacy: Context & Challenges. New-
ark, DE and Bloomington, IN: International Reading Association and ERIC, Clearing
House on Reading and Communication Skills.

PNUD (2003). Atlas do Desenvolvimento Humano no Brasil. IPEA / Fundagao Joao
Pinheiro / PNUD. Database www.pnud.org.br/atlas/instalacao/index.php ,
March, 22, 2005

PONTUAL, Pedro (1995). Desafios pedagogicos na Construgao de uma relagao de par-
ceria entre movimentos populares e o governo municipal da cidade de Sao Paulo na
gestao Luiza Erundina: A experiencia MOVA - SP 1989 - 1992. Sao Paulo: PUC.

PROGRAMA Alfabetizagao Solidaria. Alfabetizagao Solidaria: avaliagao final: segundo
semestre de 1998. Brasflia: Programa Alfabetizagao Solidaria, 1999.

RIBEIRO, Vera Masagao (org.) (2003). Letramento no Brasil. Sao Paulo: Global.

SECAD/MEC (2005) Analfabetos e Alfabetizagao de adultos (PNAD 1992-2003). Brasilia: Secretaria de Educagao Continuada, Alfabetizagao e Diversidade / MEC.

SOARES, Magda B. (1992). Literacy Assessment and its implications for Statistical
Measurement. Paris: Unesco.

SOARES, Magda B. (1998). Letramento: um tema em tres generos. Belo Horizonte:
Ceale: Autentica.

STREET, Brian (1984). Literacy in theory and practice. Cambridge: Cambridge Uni-
versity Press.

TRIBUNAL de Contas da Uniao (2002). Relatorio de auditoria Operational. Programa
Educagao de Jovens e Adultos - Alfabetizacao Solidaria. Brasilia: TCU.

UNESCO (1978). Recommendation concerning the International Standardization of
Educational Statistics. Paris: UNESCO.

WAGNER, Daniel (1999). Literacy skill retention. In: WAGNER, Daniel; VENESKY, Richard & STREET, Brian. Literacy: An international handbook. Colorado: Westiew.