7 Peasant Friendly Plant Breeding and the Early Years of the Green Revolution in Mexico

The agricultural programs collectively known as the “Green Revolution,” initiated in the 1940s with funding first from the Rockefeller Foundation and later from the Ford Foundation, were remarkably successful in some respects but disappointing in others. On the one hand, within a relatively short time various developing countries increased cereals production several fold so that imports were no longer necessary. On the other, after three decades it was clear that the programs’ declared aim to alleviate world hunger had not been realized; in some regions where Green Revolution programs had been in operation, rural poverty (effectively synonymous with hunger) actually increased. The explanation for this outcome, advanced by a number of critics during the 1970s, was that the high-yielding plant varieties and intensive cultivation techniques produced by the Green Revolution were adopted primarily by large commercial farmers. Peasant farmers, in contrast, lacked both the capital and the appropriate growing conditions, such as irrigation, necessary to take advantage of the new technology.¹

Why did the Green Revolution’s agricultural scientists pursue such an inappropriate— peasant unfriendly—form of plant breeding? Several explanations are conceivable. In view of the sharp decline in small farms in both Western Europe and the United States, especially since the 1950s, some observers might be tempted to conclude that the revolution developed a technology suitable for large commercial farms because there was simply no alternative. That is, there was no advanced technology available in the 1940s and 1950s that could have boosted the productivity of small farms. This explanation is untenable, as technology suitable for peasant farms was developed during the green revolutions that started in Western Europe during the late nineteenth century. For example, in German speaking Europe several states established plant-breeding stations around 1900 that explicitly aimed to bring the benefits of advanced breeding technology to the peasant farmers who predominated in those regions, and the evidence indicates that these efforts were successful in invigorating regional agricultural economies. One consequence of such policies in Germany was that the only size-category of farms to increase (both in number and in proportion of the total acreage) between the late nineteenth century and the 1960s was that of ten to twenty hectares (twentyfive to fifty acres). During the 1940s and 1950s when the classic foundation-sponsored Green Revolutions for the developing world were being planned, therefore, a successful European model for promoting small-farm productivity was available.²

Were the Green Revolution’s designers simply unaware of European policies, perhaps seeing them as irrelevant for American conditions where farms were on average far larger? Or were they familiar with European developments but took the view—sometimes expressed by US secretaries of agriculture during the 1950s and 1960s—that small farms were not worth bothering with since they could never be as efficient as large ones? Were they aware of the European approach but preferred to take the path of least resistance, thinking that it would simply be easier to work with large farmers who tend to embrace new cultivation methods? Or were foundation officials worried that targeting peasant farmers for assistance might be seen by host governments as an unwelcome “political” intervention?³

These questions can be answered by looking at the early history of the MAP, the first of the Green Revolution programs undertaken by the Rockefeller Foundation. Most historians of the MAP tend to emphasize the extent to which the program’s advisors and staff relied upon an American model of agricultural development that was largely inappropriate for Mexican conditions or have questioned the MAP’s commitment to alleviating rural poverty and hunger. While this is probably an accurate characterization of the program in the 1950s, it does justice neither to the original design of the program in the early 1940s nor to the initial aims of its maize-breeding work. Thus, the program underwent a substantial shift between the 1940s and 1950s.⁴

In fact, the foundation’s agricultural officials were almost certainly aware of European developments, and some of them, as well as advisors to the MAP, also had first-hand experience of the problems faced by small farmers in the United States and elsewhere during the interwar period. Furthermore, perhaps surprisingly, the foundation’s declared aim of alleviating rural poverty was not just posturing by an organization anxious to be seen as an agent of philanthropy; some foundation officials and advisors appear to have been genuinely concerned to improve rural living standards through increasing the productivity of peasant farms. In addition, the program’s early work in maize breeding indicates that rather than just trying to develop American-style hybrid maize varieties for Mexican conditions, the breeders used quicker methods to produce varieties more appropriate for the circumstances of poor farmers. Once the MAP had been running for a few years, however, it became apparent that the task of getting new varieties and cultivation practices to small farmers, though urgent, was going to be difficult, not least because the MAP possessed neither the facilities nor the formal authority to undertake this task on its own. Faced with the need to make some kind of impact quickly, MAP staff chose to concentrate upon projects that were likely to find a ready audience. This meant setting aside the needs of peasant farmers to develop high-yielding varieties especially suited to large commercial farms. In effect, the program was abandoning its original aim of alleviating rural poverty.

The Rockefeller Foundation’s interest in agricultural development did not begin with Mexico nor were its agricultural advisors unaware of the problems of peasant agriculture. Before the First World War, for example, the General Education Board (GEB) had funded extension programs in the American South, and during the 1920s the International Education Board’s (IEB) support for European higher education had included agriculture. Among the agricultural experts associated with the IEB the most important was A.R. Mann. Dean of Cornell University’s School of Agriculture, Mann was the board’s director of agriculture from 1924 to 1926; and during the 1920s he visited the major European experiment stations on its behalf, alert to the possibility that American agriculturists might have something to learn from them. Among the sites he visited were German plant-breeding stations, and although his diaries do not single out their peasant-friendly orientation for comment, he seems to have been aware that European peasants had different needs from most American farmers.⁵

Unlike some of the IEB’s other advisors, Mann was not single-mindedly in favor of supporting “fundamental research” in the agricultural sciences. Foundations, in his view, could pursue two quite different paths toward improving production. One of these was to fund work aimed at longer-term improvement, the other was to support “practical measures” that would convey best practice to farmers, as the GEB had done before 1914.⁶

This practical emphasis was also evident in the foundation’s support for agricultural development in China. In 1924, for example, Mann agreed to provide IEB funding to support a “Cooperative Plant Improvement Program,” which would send breeders from Cornell University to the University of Nanking. The program was tailored in a number of respects to the needs of small farmers. The Cornellians, for example, ignored the requests of Chinese cotton-mill owners and chose to focus instead on the major staple crops, among them sorghum because of its ability to withstand drought. And they chose to use selection rather than hybridization because it was cheaper and faster. Part of the program entailed funding Chinese students to study plant breeding, but an American agricultural scientist advised the board that it would be a bad idea to send young Chinese to study in the United States because they would come back with “an American training that doesn’t fit Chinese conditions.” The conditions in question included the fact that average farm size was only 3.5 acres, and the breeders realized that this made Chinese peasants necessarily cautious about adopting new varieties (unlike their American counterparts who had land to spare for risky trials). This meant that extension was crucial—a point that the breeders eventually recognized. Though the program’s breeding work managed to acknowledge peasant needs, however, in one historian’s view breeders never fully solved the dilemma of extension. To what extent Rockefeller agricultural advisors learned a lesson from this experience is not clear, but it is interesting that in its subsequent China Program between 1935 and 1937 the foundation took the view that the program’s agricultural work should focus upon the application of existing knowledge and its dissemination rather than upon research.⁷

This emphasis upon “practical measures” and extension may have characterized the GEB’s Southern Agriculture Program, which Mann directed from 1936. Before he took over the program, it was concerned with public education, “negro education,” and (from 1933) with southern white colleges and universities. In 1936, however, the program’s declared aim was to strengthen those areas of education and research that dealt with the economic and social problems of rural areas.⁸

After the end of the IEB in the late 1920s, Mann continued to play a major role in Rockefeller agricultural programs, initially as director of the Southern Agriculture Program and then as vice-president of the foundation. It was natural, therefore, that he was consulted on the design of the Mexican program, and from 1946 he served as part-time officer in the Division of Natural Sciences with responsibility for the MAP. It is thus largely through him that the foundation’s pre-war experience in agriculture was brought to bear upon the MAP. As planning for the program got underway, a Survey Committee (later the Advisory Committee for Mexican Agriculture) was also set up late in 1941–consisting of a plant pathologist (Elvin Stakman), a plant breeder (Paul Mangelsdorf), and a soil scientist (Richard Bradfield). These three, along with Mann, effectively designed the MAP.⁹

Although the publicly declared aim of the MAP was to alleviate poverty and hunger, this does not mean that it was necessarily prompted by humanitarian concerns. Indeed, officers and advisors on occasion attributed a strategic political significance to the program:

Communism makes attractive promises to underfed peoples; democracy must not only promise as much, but must deliver more… . Asiatic and other underprivileged people attribute their present plight to the domination of the capitalist colonial system… . In this struggle for the minds of men the side that best helps satisfy man’s primary needs for food, clothing and shelter is likely to win… . Appropriate action now may help [people of developing nations] to attain by evolution the improvements, including those in agriculture, which otherwise may have to come by revolution.

Additionally, foundation officials were aware that the early 1940s was a good time in which to launch an aid program in Mexico since the newly elected Mexican government appeared more interested in cooperating with the United States than its predecessor (which had nationalized the oil industry without compensation).¹⁰

But how serious were officials and advisors about reducing rural poverty and hunger in Mexico? Considering the general aims of the original MAP, it appears that those involved agreed that it was important, not simply to boost agricultural production overall (which could have been achieved by targeting assistance just to large commercial farmers), but also to alleviate hunger and low standards of living among Mexican small-holders. Indeed, the original suggestion to establish an agricultural program in Mexico came from staff in the Rockefeller’s International Health Division, who thought that it would usefully complement their existing public health program in Mexico. This idea of integrating public health with agricultural development—an early example of what was later regarded as best practice in development policy—found favor with the Survey Commission in 1941, with Warren Weaver in 1946, and with Stakman in the early 1950s although it seems not to have been implemented.¹¹

In addition, program designers were aware that alleviating rural poverty would mean addressing the specific needs of peasant farmers and would require more than just the use of commercial inputs. Both scientists and officials were clear, for example, that it would not be possible merely to apply established American cultivation methods to Mexican farming. In its report to the foundation following the first exploratory visit to Mexico in 1941, for example, the Survey Commission noted that they had assessed the state of Mexican agriculture: “Not solely by American standards but in the light of the history and traditions of the Mexican people. It would be wholly impossible, even if desirable, to impose the modern American culture upon Mexico; any improvement that is to be brought about must come within the framework of Mexican culture.” When officials consulted Carl Sauer, a social scientist with extensive experience of Mexico, they got the same message:

A good aggressive bunch of American agronomists and plant-breeders could ruin the native resources for good and all by pushing their American commercial stocks… . Mexican agriculture cannot be pointed toward standardization on a few commercial types without upsetting native economy and culture hopelessly. The example of Iowa [i.e., where the maize crop was based almost entirely upon a few hybrid varieties] is about the most dangerous of all for Mexico. Unless the Americans understand that, they’d better keep out of this country entirely.¹²

While the advisors’ awareness of the complexity of the challenge facing the MAP is significant, perhaps more striking are the specific measures that the program adopted in order to improve Mexican agriculture. To be sure, it does not seem that Warren Weaver (as director of the Division of Natural Sciences that oversaw the MAP), Mann, or the Advisory Committee consciously considered European models in designing the MAP, although Stakman’s ties with German plant breeders meant that he would almost certainly have heard about the peasant friendly stations there. Nevertheless, several features of the initial breeding program were peasant friendly in the sense that they were designed with the needs of small farmers in mind, in particular resource-poor farmers who could not afford to purchase new seed annually. In view of the criticism later directed at the MAP for giving wheat breeding such a high profile, it is significant that at the outset the advisors were agreed that maize was the crop to focus upon. Since it was central to most Mexicans’ diet and was grown by the great majority of small farmers, maize offered “the greatest opportunity for improvement in Mexican agriculture.” Equally important, advisors were not single-mindedly bent upon developing new high-yielding varieties (as had been done in the United States). The most pressing problem, they argued in 1941, was to improve cultivation practices (because of serious erosion and depleted soils). Improving varieties was the next most important issue, but the advisors envisioned these being provided via “introduction, selection or breeding [my emphasis].” Accordingly, the MAP initially devoted considerable effort to testing existing Mexican varieties of maize to identify the best one for each locality. Simply by redistributing these to the most suitable locality, one advisor recommended, the quality of maize cultivation could be improved quickly without having to wait years for the development of new varieties. This approach was richly rewarded when it was discovered that yield in some regions could be increased by 20 to 30 percent by introducing a variety native to another region.¹³

In addition, as Karin Matchett’s study of the MAP has shown, the program’s advisors took into account the economic circumstances of peasant farmers. As the Survey Commission observed in 1941, the problem with US-style hybrid maize was that seed had to be purchased each year, “and the small farmer in Mexico has neither the cash nor the initiative to do this.” The breeder on the commission, Mangelsdorf, pointed out that hybrid maize varieties had not been successful in the American South where conditions were similar to those in Mexico. His experience in East Texas had been that it was hard to get small farmers there to purchase hybrid maize seed every year instead of saving seed from the previous harvest. Where farms were small and maize was grown for subsistence rather than for sale on the market, therefore, it was necessary to develop improved varieties that could be replanted year after year. Open-pollinated varieties—where crossing occurs spontaneously in nature rather than under the control of the breeder—he thought, met this need while hybrid maize did not.¹⁴

According to Matchett, although Edwin Wellhausen’s maize-breeding program at the MAP included constructing conventional hybrids using the double-cross method worked out in the United States, most of his efforts through the 1950s used quite different ways of improving maize. Some of the work relied on a simple and traditional method (mass selection) to improve Mexican landraces, but the majority of it was dedicated to making synthetic varieties, a kind of quick-and-dirty hybrid that was relatively high-yielding and whose seed could be replanted each season. Part of the rationale for this was agro-ecological. As a MAP progress report in 1944 noted, conventional hybrid maize varieties in the United States were so heavily tailored for a particular region that “they are complete failures elsewhere while inferior open pollinated lines are more adaptable and can be grown in various regions.” But the MAP’s breeding strategy was also pragmatic; it was much quicker to produce synthetic varieties than double-cross hybrids (which could take over ten years), and, as Mangelsdorf emphasized, Mexico was a good place to turn out something in a hurry because almost anything would be a substantial improvement. The initial results were promising; the first synthetic varieties released by the MAP in 1948 yielded about 30 percent more than the benchmark variety.¹⁵

Despite the MAP’s promising start with a peasant-centered approach to development, it soon became evident that improving the state of peasant agriculture was not going to be easy. By the late 1940s there were signs of disagreement—among advisors as well as within the foundation—on the most effective strategy for transforming Mexican agriculture. And by the 1950s the MAP was no longer pursuing some of its original aims.

At the center of this shift was the question of what role the MAP should adopt in relation to extension. Several months before the Survey Commission’s visit to Mexico, Mann had recommended that a Mexican program should take a two-pronged approach, pursuing not only research but also extension, since the latter offered the promise of a relatively quick impact. Following their visit in the summer of 1941, similarly, it was clear to the Survey Commission that an immediate improvement to Mexican agriculture did not require the production of new practices or varieties. An enormous improvement in yield could be achieved just with the application of existing knowledge. During his visit to Mexico in 1943, Bradfield was struck by the fact that basic forms of good practice such as crop rotation or fertilizing with manure were rare, and the wooden plow was almost universal. The highest priority, he stated in his report to the foundation, should be extension. Following another visit two years later, he reiterated his concern; something had to be done to improve soil fertility, he argued, before the full fruits of the work on plant disease and new varieties could be exploited. The time was approaching, “if we have not already reached it, when we should begin to think about how the information obtained in the research program can be made most effective in Mexican agriculture.” In another year or two, he thought, there would be enough knowledge upon which to base an extension program. At this point the other members of the Advisory Committee did not agree, arguing that the MAP should stick to research for the time being. But after their visit to Mexico the following year (1946), Mangelsdorf and Stakman concurred with Bradfield: the MAP’s research had been so successful that it was time for the foundation to start pressing the ministry for the development of a properly designed extension program or perhaps even to become involved in extension itself.¹⁶

This initial emphasis upon extension is significant because it reflected the advisors’ concern to reach small farmers. Large farmers, in Mexico as elsewhere, were much better placed to look after themselves. Unlike their smaller brethren, they could afford to take risks with new methods and had the capital to invest in them; many were thus keen to cooperate with the MAP by offering land for field trials and adopting new varieties. But to disseminate the requisite knowledge to small farmers required a functional extension service, which Mexico did not yet have. The existing service, Bradfield pointed out, was totally unsatisfactory. Since its staff had no means of transportation, they never came into direct contact with farmers and were thus reduced to distributing leaflets and answering letters. Given low levels of literacy among peasants, this was not a viable way of reaching the vast majority of farmers. Nor was the MAP ever likely to have either the resources or the manpower to take on such a huge job itself.¹⁷

By 1946 the advisors were not the only ones to have reached this conclusion. During their own visit to Mexico that year, Weaver and Mann met the minister of agriculture who expressed the view that it was time to begin extending the findings and new seeds to farmers. The minister was keen for the MAP to take the initiative in developing a Mexican extension service, and the Advisory Committee concurred, outlining how such a service–funded and administered by government–might be organized. Weaver and Mann endorsed this new emphasis: “a start should be made now in the difficult but absolutely essential business of introducing into the actual agricultural practices of Mexico the improved materials and methods being developed. We can’t finance an extension system for Mexico, nor is Mexico ready for such, but extension must anyway be started.” Within a few weeks of returning from Mexico, Weaver had drawn up a list of matters to discuss with MAP Director George Harrar, among them extension. Harrar agreed with the Advisory Committee that the program should have an additional staff member who would concentrate upon extension and liaise with Mexican officials on the development of the service. And it was agreed that Harrar would get the ball rolling by writing to the minister and explaining precisely what, in the MAP’s view, the Ministry would need to do in order to get the service underway.¹⁸

In the event, things developed slowly. In 1947 the MAP held farm demonstrations in various regions, and in 1948 the program appointed Mortimer Barrus to work with the Ministry on designing the extension service. But by 1949 Barrus was not receiving the cooperation he wanted from Ministry staff, and when the foundation declined to intervene on his behalf, he resigned. Thereafter formation of the service dragged along more slowly than foundation officials or experts had hoped, and the MAP’s collaboration with the Ministry remained difficult. By the early1950s there was still concern in the foundation that the MAP’s considerable technical advances were not being made available to the great majority of Mexican farmers, and some advisors and program staff were clearly frustrated. Reporting on his site visit in 1953, Bradfield noted that, while in principle the use of mineral fertilizers or crop rotation using alfalfa would enormously improve Mexican agriculture, in practice “to handle a system of farming of this more complex type…will require more managerial ability and more capital than many Mexican farmers have at the present time.” But meeting those needs required an extension service and improved credit arrangements. The dilemma with irrigation was similar. As Wellhausen (who became director of the program in 1952) later remarked, in areas with adequate rainfall it was clear that by simply applying existing knowledge farmers could produce far more than they currently did. But since water was not being efficiently used, the (predominantly small) farmers who needed it were not receiving it, and how to increase yield in areas of low rainfall was much less clear. Although the MAP was adept at developing technical solutions, only government could provide the necessary infrastructure: “Mexico needs to do something about increasing production in these marginal corn-growing areas because it is in these areas where the population is also beginning to increase… The people from these areas are beginning to march on the cities and form the slums.”¹⁹

Some historians have been critical of the MAP’s failure to develop an effective extension program, arguing that its staff “seemed to feel that large grassroots [extension] campaigns were not part of their agenda,” an attitude that allowed them to “bypass the vexing problems of rural poverty.” One problem with this critique is that it is difficult to imagine how a program as small as the MAP during the 1940s—with relatively limited resources and half a dozen scientific staff—could have mounted anything but a token extension effort. More fundamentally, what the criticism overlooks is the extent to which MAP staff and advisors were aware during the early years of the program that extension was essential and that it would be necessary to assist the Mexican government in strengthening the existing service.²⁰

The fact that the MAP’s work failed to reach peasant farmers was not only due to the shortcomings of the Mexican extension system. For, although Weaver and Mann had called early on for more attention to be given to studying the “economic matrix in which the scientific agricultural studies are placed,” the MAP was slow to move in that direction. One of those who backed Weaver and Mann’s call was William I. Myers, professor of agricultural economics at Cornell and a trustee of the foundation since 1941. Soon after the MAP started, he had spoken with Harrar and Weaver about the need to bring in agricultural economics to supplement the biological sciences and make the MAP a more rounded research program, but with little success. For one thing, the natural sciences and social sciences were situated in different divisions of the foundation and had no experience of collaboration. For another, despite Weaver’s verbal support for studies of economic context, he occasionally expressed the view-to Myers’s annoyance-that farming was “just applied biology.” The MAP staff had been trained in biological sciences with little exposure to either economics or rural sociology, and “they were suspicious of what they didn’t know.” Myers kept pushing. Responding to a MAP progress report in 1951, he remarked that although he liked the report, it was optimistic on what remained to be done. “We have not yet made even a beginning in the study of economic and social problems that are also important in improving the general level of well-being of the countries concerned.” The foundation, he thought, could research such problems and suggest methods of solving them. By this time even Weaver was coming around to the idea that more work on economic questions was needed, writing to the Advisory Committee that he thought future grants for agriculture should include not only scientific work but “studies of a less specifically scientific nature but also addressed to long-range limiting problems of agriculture (water, land tenure, taxation, etc.).” But Weaver was either unwilling or unable to push this idea since nothing happened for several years. When Stakman recommended in 1954 that the MAP hire a rural sociologist and an anthropologist, one foundation official agreed but thought that such staff only needed to be added in the course of the next decade. A social science perspective on agricultural development was evidently not a priority within the foundation, and it was not until 1956 that an agricultural economist was finally added to the staff.²¹

Why, then, did the MAP give up its original aim of raising productivity on peasant farms? That is the crucial question, but the answer is undoubtedly too complex to develop in depth here. What is worth doing, however, is to outline a number of hypotheses that emerge from the archive and that merit further consideration by historians.

While the views and recommendations of foundation officials were important, the MAP’s actions obviously depended as well upon the field decisions of its own staff, above all, Harrar. And there are indications that Harrar did not always see eye to eye with either officials or advisors. In the autumn of 1946, for example, when Weaver and Mann had returned from Mexico, persuaded that more attention should be given to the economic context in which scientific agriculture was practiced, Harrar objected that adding an agricultural economist to the program would be “dangerous.” His view may have prevailed, since it was another decade—by which time Harrar was no longer director—before an agricultural economist was finally appointed. Similarly, although the MAP’s original focus was on maize, beans, and wheat, by the spring of 1946 Harrar was interested in expanding the scope of the program to include crops with export potential such as fruit, vegetables, sugarcane, oil-bearing plants, pharmaceutical plants, and rubber. Whether he conveyed this proposal to the minister of agriculture is not clear, but it may be significant that when the minister met with Weaver and Mann later that year, he emphasized that the most important tasks for the MAP were maize, beans, and wheat; other crops could be taken up later. Why might Harrar have deviated from the general line adopted by officials and advisors? There are hints that personality may have played a role. Mangelsdorf, for example, said that “There is some danger that a man of Dr. Harrar’s temperament, eager to get things done, and constantly confronted with immediate problems, will tend to lose sight of the long-time objectives of the program.” There was certainly no shortage of “immediate problems.” As Stakman later recollected, staff were inundated right from the start with people wanting them to work on particular projects that had nothing to do with the main food crops (e.g., improving limes, vanilla, and coffee). If the foundation granted Harrar considerable operational freedom–and there are signs that it did–his conception of what the program should do may have diverted its work away from the direction envisioned by the advisors.²²

The problems in implementing the original vision of the MAP, however, were not confined to matters of governance. For the program depended on the Agriculture Ministry’s extension service to disseminate improved cultivation practices to the peasantry. And a change of government in 1947 brought a different policy on extension, with the result that Harrar’s efforts to persuade the new leaders of its importance ran into “many obstacles and a good deal of resistance.” A service was finally established in 1953, but even by the mid-1960s it had not developed as quickly or satisfactorily as he would have liked. The most serious obstacle from the program’s point of view centered on the mechanisms for the distribution of seed from the MAP’s improved maize varieties. As early as 1946 the Advisory Committee was concerned about the government’s arrangements for multiplying and distributing the new seed. Although Stakman conceded that ideally a state institution should distribute the seed, in practice he and the others worried that the proposed new system was open to abuse and that seed would be made available to large farmers rather than to small ones. The solution, he suggested, might be for the MAP to distribute its own seed (although how its staff could have managed such a gigantic task is not clear). But his suggestion was not taken up, and by the spring of 1947 Stakman was disturbed to see that responsibility for distribution had been assigned to the newly established National Corn Commission. The MAP, he reckoned, would have to make the best of the situation and try to prevent the same thing from happening with seed from other crops.²³

The resistance that the MAP encountered while trying to improve the extension service is not surprising, given the Mexican economic and political situation after 1940. As de Alcantara, Lewontin, and others have argued, the program of rural development championed by the left-inclined government of Lazaro Cardenas between 1934 and 1940 was abandoned by the center/right governments that came to power over the next twelve years. The latter’s power base consisted of an alliance between urban businessmen and large landowners who agreed that public funding should be channeled toward “progressive” commercial farms rather than “backward” peasants (through irrigation works, subsidizing wheat prices, and generous credit terms for big farms), and the large agricultural surpluses thus generated would provide capital for the industrializing economy. Since large farmers received the technical information and assistance they needed, only a rudimentary extension service was required. Thus “post-Cardenas governments gave extremely low priority to the kinds of programs required to support production within commercial ejidos [peasant communes].”²⁴

Unfortunately the limitations of the extension service, as MAP staff saw it, derived not only from a lack of government enthusiasm; cooperation with Mexican scientists was also hindered by professional rivalry as well as a clash of cultures. On the one hand, Mexican agricultural scientists appear to have been sensitive about well-funded foreign scientists—especially from north of the border—arriving at the Ministry and questioning traditional ways of doing things. Some of the maize breeders, for example, did not want the MAP’s new varieties to displace older ones. On the other hand, the Mexicans’ notions of professional status hindered good working relations with peasant farmers. From the visitors’ perspective, Mexican agricultural expertise was “book knowledge,” not grounded in hands-on experience. Research that took the scientist into the field–something that MAP staff took for granted–enjoyed little status, and even younger Mexicans thought that fieldwork was beneath them. That most Mexican scientists had little sympathy with peasants and were inclined to lecture them only aggravated this problem.²⁵

Finally, the MAP had to deal with political issues. As representatives of an American foundation-and especially one with connections to the oil industry–MAP staff had to tread lightly when offering advice that had policy ramifications. One such issue was land reform. Following the revolution of 1910, peasants had been given tracts of two to four hectares each (five to ten acres), and this farm size remained enshrined in land reform policy through the 1940s. But MAP experts believed that such small farms only could allow subsistence but no surplus production for the market and thus no purchasing power for the peasant. It would be better, they concluded, for land reform in future to parcel out larger tracts. But to voice such views publicly might have jeopardized the program. As Myers later noted, something of this kind had occurred in the Philippines where a western agricultural economist had criticized the landholding system for preserving many large farms, provoking an outcry that nearly shut down the development program. In Myers’s view, the economist’s analysis was actually correct, but: “You don’t go into a country … no matter how much you know and how bad their situation is–and tell them what’s wrong and what ought to be done about it…. [The problem of large landholdings] is one of the basic problems of Latin America–but you can’t solve it when you land in the country–[and] tell them how lousy their system is.” Moreover, the history of US intervention, both military and economic, as Harrar recognized, made Mexicans suspicious of American motives, extending even to plans for the reform of Mexican agricultural education. Besides keeping their mouths shut on land reform, MAP staff tended to stay clear of the US embassy because, as Stakman recollected, “I think at that time the Foundation felt that they should retain their independence of action and avoid any appearance of being political.”²⁶

Faced with the scale and complexity of the problem of disseminating MAP results to small farmers, the MAP’s lack of control over the extension service, and the need to steer clear of offering advice on basic questions of agricultural policy, there are signs that the staff began to concentrate upon problems where they thought they could make progress. In 1947, for example, Wellhausen argued that areas where a maize surplus could be produced should be addressed first; subsistence areas could be left until later. And in 1951 Bradfield recommended to the Advisory Committee that: “Since the improvement of agriculture is dependent upon improvements in education, health, transport and the availability of [inputs], priority should be given to the few situations where a well-rounded program of development seems most probably [sic] over situations which are not yet so ready.” Similarly, when asked in the 1960s how maize yield could be increased in areas of low rainfall, Wellhausen replied, “I don’t really know. These areas we have not been too concerned about. We’ve … concentrated our efforts on the areas which are more productive from the standpoint of corn.” This tendency to take the path of least resistance may help to account for the effort eventually devoted to wheat breeding. For one thing, the Mexican government was keen to boost wheat production to meet a growing urban demand and reduce imports, but extension problems were also easier with wheat since growers were literate and had the capital to take advantage of intensive methods. As Myers later suggested, the rapid success of the wheat program was so spectacular—Mexican output trebled in only ten years and by 1958 the country was exporting wheat—that “probably they haven’t given adequate attention to other things” such as farm management. It would appear, therefore, that work on easier problems aimed at large commercial farmers gradually displaced that designed to help peasants.²⁷

To conclude, Rockefeller experts and officials believed during the early years of the MAP that the program could and should make an impact upon Mexican poverty and hunger. To this end, they endorsed a program of extension, varietal-testing, and breeding designed to help the small farmer. By the 1950s, however, the program had taken on a rather different cast. Just why the original vision became derailed is not clear but may include factors such as the director’s own convictions as to how the program should develop, difficulties in collaborating with Mexican experts, and a lack of control over key institutions such as the extension service. The resulting modus vivendi was a division of labor in which the MAP concentrated upon research and training, while leaving responsibility for extension to the Ministry. And, although the success of their research in alleviating hunger was ultimately dependent upon economic and political policies that established the framework in which the new technology would have to operate, this policy arena was one that MAP staff and foundation officials sought to avoid. As a result, the program’s greatest impact, despite its original intentions, was upon large farms. For some this remained a disappointment. When the Advisory Committee returned to Mexico in 1962, one of the things that disturbed them was to see that, despite the useful knowledge and practices the MAP had generated, the lot of the Mexican peasantry remained largely unchanged. As Mangelsdorf remarked, “Some of this has [also] happened in the US … I don’t know what the answer is.”²⁸

Demonstrating that the MAP was a dynamic entity, capable of shifting in response to circumstances, opens up a new area of inquiry for future studies, not only of the MAP but also of other green revolutions. As the first of these programs, the MAP was important as a model for subsequent programs in Latin America and elsewhere, and it remains to be seen whether the later programs succeeded in learning from the MAP’s experience or blindly reproduced it.

7.1 Acknowledgments

This paper was written in part while I was a visiting scholar at the Max Planck Institute for the History of Science in Berlin, and I am grateful to Hans-Jörg Rheinberger and to Institute librarians for their support. An early draft of the paper was presented at the meeting of the Society for the History of Technology at Washington, DC, 2007. I thank participants at that session, as well as my colleague, Inderjeet Parmar, for critical comments.

7.2 References

Footnotes