«Item type text; Electronic Dissertation Authors DeJong, David Henry Publisher The University of Arizona. Rights Copyright © is held by the author. ...»
Analysis of the adjudication maps requires some explanation.614 With field data for the reservation digitized, and with aggregate data for all farms in the state available from the U.S. Census Bureau, I could compare crop pattern differences.615 Crop patterns provide a measurement of the health and vitality of the reservation economy and whether or not reservation-based agriculture was on par with, or on the periphery of, Arizona agriculture in 1914.616 Differences in mean village field sizes on the reservation provided the data needed to determine inter-village differences and the effects of water loss.
There are several caveats to consider before analyzing and explaining the survey maps. Southworth spent six months (January through June) in the field conducting the adjudication survey.617 This timeframe is significant in that, under normal conditions, it might be skewed toward winter crops with an accounting of only some summer crops (i.e., fields surveyed in May or June). While summer crops would have been planted and winter crops would be nearing maturation in some areas (Santa Cruz, Blackwater, Gila Crossing, Cooperative and Maricopa), the fact that not all villages were surveyed during the spring precluded some from being adequately represented on the survey map, particularly Sacaton, Santan and Casa Blanca. Since Pima agriculture was traditionally 614 I am indebted to Dr. Wendy Bigler of Southern Illinois University at Carbondale Geography Department for the GIS data on the 1914 Southworth plain table maps. Dr. Bigler and I spent several days in the summer of 2004 examining maps in the San Carlos Irrigation Project archives in Coolidge, Arizona, and preparing to digitize them.
615 Crop patterns for the state of Arizona are taken from the Sixteenth Census of the United States: Agriculture, Volume I, First and Second Series State Reports, Part 6, Statistics for Counties (Washington, DC: GPO, 1942), pp. 384-94.
616 This seems to be an important consideration when we recognize the Indian Service was prepared to allot the reservation. If the Indian Service were seeking to provide an “equal chance” for the Pima then it would be incumbent upon the department to ensure the Indians were actually in a position to compete with non-Indians, as its rhetoric implied. This, of course, was rarely the case and the Indian Service seemed less concerned with competitive Pima farmers than it did with allotment.
617 Santan and Casa Blanca were surveyed between January and March; Sacaton in February; Santa Cruz and Blackwater in April;
Gila Crossing between April and May; Cooperative in May; and Maricopa in May and June.
262 seasonal, with grains the dominant winter crop and cotton, corn and squash more dominant in the summer, the data may be skewed towards the former. This limitation, however, is mitigated by the fact that insufficient water limited most fields to one crop per year, as Pima farmers Juan Thomas, Havelena and John Makil explained in 1914.618 A second factor is that off-reservation crop data for the State of Arizona, while reflecting an accurate picture of all crops grown, provide values taken from the 1910 and 1920 U.S. Census Bureau crop census. The State of Arizona did not collect crop statistics in 1914. Consequently, local cropping patterns (i.e., Pinal and Maricopa counties) cannot be directly compared with reservation crop patterns, which would have been the ideal comparison. To address this challenge, I used averaged data from 1910 and 1920 to reflect the state of agriculture in Arizona in 1914. Census data, however, limits crop patterns to total acres and proportions only, not specific field (or farm) sizes and number of fields planted to specific crops, something the Pima adjudication survey does provide.
The present analysis also dropped any consideration of fruit acreage in Arizona due to limited data availability. Because the Southworth data are for fields (not necessarily farms) and the Arizona data are for farms (not fields) no direct comparison can be made of farm sizes, although the average of the former was estimated by Southworth at between ten and fifteen acres, while the latter was 135.1 acres.
A third consideration is the manner in which the adjudication maps refer to fields.
Southworth lists each field as planted to a single crop when this was more than likely not the case. Pima farmers frequently planted “a portion of each field … to pumpkins,
squash, and melons” each year. Other smaller plots of vegetables would also have been planted but fail to show up in the survey. Fruit production was minimal and, as Reed explained, would have been “confined to small orchards in connection with individual farms.” This might explain why there were only eleven fields marked “orchards” out of the 3,178 fields. While the map scale is detailed and shows fields smaller than one acre, Southworth omitted small garden plots or simply consolidated them with the more dominant crop planted in the field.619
A final consideration is the variance of crop selection from one village cluster to another. Two clusters stand out: Sacaton and Santan. These village clusters were the most diverse in terms of cropping patterns (see table 6), likely a reflection of the U.S.
Department of Agriculture having a field research station in Sacaton, which was just across the river from the Santan district. Farmers in these clusters would have had greater access to markets and experimental crop seeds than those in more remote village clusters.
These latter clusters grew a greater percentage of grains, as shown in table 9. Sacaton was also the agency center with a bustle of government activity and influences from the agency farmer, and Santan was the recipient of the first federal reclamation project on the reservation.
To compare cropping patterns on the reservation with those off the reservation required the construction of a graph (not shown) to display the value of each crop. Since this pointed only to the difference between the reservation and the state, I calculated proportional values for the reservation based on a per capita distribution of crops across the state. Population data was secured from the 1910 U.S. census for Arizona and a 1904 Indian Service census for the reservation. Table 7 shows the actual and proportionallycalculated values.
The bar chart visually illustrates the differences in cropping patterns. Based on population and geographic area, Arizona was expected to have significantly larger crop acreage and greater totals by crop. However, when an examination of Pima actual (what was cultivated in 1914) and Pima proportionally-calculated (what should have been grown based on a ratio of acres to population) values is considered, important differences appear. Based on an average ratio, I expected the reservation to have an increased acreage of hay, cotton and other crops (vegetables), a decreased acreage of corn, and 82% less grain. In other words, based on a state-wide ratio, average need and crop selection, the Pima should have grown more fiber crops and fewer food crops.
This disparity lends support to the belief that Pima farmers expended more energy on growing grain (food) crops since they would not only provide subsistence for the people but they would also require the least amount of water. Pima farmer George Pablo supported this hypothesis when he reported that some Pima farmers fed grain to their cattle in 1914, “after convincing themselves that there will be no maturity of their planted
and this alkaline seepage water” caused grain not to mature or reduced its yield.620 Clearly, conditions were not conducive to Pima agriculture.621 Table 9: Proportion of Fields Sown to Grain by Acres based on GIS Analysis
To develop further the theory that disproportionate grain production might have been a product of insufficient water, I also tabulated the proportion of crops grown by village clusters, as shown in table 9. Analyzing cropping distribution by village clusters points to several patterns. First, there were clear shifts in a number of villages and scores of farms by 1914, as whole groups of Indians simply abandoned one location and sought to reestablish themselves in another. Overall prospects for water to irrigate crops were unpredictable at best, leading farmers to abandon their fields or plant fewer acres. Many 620 Statement of George Pablo, p. 31. Statement of Ho-Ke Wilson, p. 5.
621 Using the cropping data shown in table 8, a chi-square test for goodness-of-fit was performed. When comparing the observed values of individual crops such as hay, corn, cotton, grain and other (i.e. garden) crops on the reservation with the proportionallycalculated reservation counts, the test result is P (χ2 13,594.2) = 0. When a chi-square test for goodness-of-fit is run on two simple crop categories (food and fiber) using the same values as above the result is P (χ2 10,700.05) = 0. The chi-square tests indicate the probability of seeing this disproportionate ratio of cropping patterns is zero given that all other factors are equal, illustrating the remarkable disparity of crop patterns on and off the reservation. In this case, the P-value of zero says that if the distribution of the actual cropping patterns on the Pima Reservation were in accordance with the proportionally-calculated value, an observed chi-square value of 13,594.2 (and 10,700.05) would occur zero percent of the time. On the former test, there were four degrees of freedom and on
the latter, there was one degree of freedom. When the test was run against the state crop values the numbers were even more dramatic:
for all crops individually the test is P (χ2 284,800.34) = 0 and when the data is packaged in two categories (food and fiber) the test is P (χ2 284,654.71) = 0.
267 families left their villages and moved to places where seepage water was available. Fields in the central and eastern—traditionally farmed—areas of the reservation were abandoned due to a dry river channel. This included the villages of Snaketown, Old Stotonic, Mount Top (part of the Casa Blanca cluster) and North Blackwater. Thirtyseven percent of the fields in Casa Blanca were fallow, as were 38% of those in Blackwater. Some farmers were no longer able to grow sufficient food for their needs.622 The overall proportion of grain per village cluster was significantly higher than the statewide proportion, as shown in table 9. This may reflect a cultural affinity for growing grain as a staple food, although Reed believed it was attributed to a lack of water. “Wheat and barley are the staple crops,” Reed informed the House committee.
“While the grains are the least profitable, yet they require the least water for irrigation and this consideration is responsible for the selection of these particular crops.” Reed further noted the Pima grew some corn and garden produce but such crops were “not a safe proposition under the gravity [irrigation] system” due to the inability to get water on the crops at the right time. “When [the farmer] is supplied with water,” Reed concluded, he “will make a living and a surplus.”623 To understand better the effects of water loss, an examination of changes in field size is necessary. Calculating cultivated mean field size by village cluster yields surprising results. Larger mean field sizes are found on the eastern half of the reservation.
Since water was the primary limiting factor, it might be assumed that larger cultivated fields would have been located in areas where seepage water was more likely to be
available or in newly established villages, such as those on the downstream portion of the reservation. However, this is not the case. The largest cultivated mean field sizes in acres (6.92 in Santan and 6.86 in Casa Blanca) were in areas where the river was dry. In areas where the river still contained water (albeit minimal amounts), mean field sizes in acres were the smallest (3.36 in Santa Cruz, 3.93 in Gila Crossing and 4.07 in Cooperative).
Blackwater (6.75 mean field size) was an exception to this rule. While the river was generally dry there, seepage water allowed crops—especially grain (92.3% of all Blackwater acres)—to grow. Maricopa, served by the Salt River (which continued to flow after the Gila River went dry), had a moderate mean field size in acres of 5.81, as shown in table 10.
When all fields are analyzed (including abandoned, not cultivated and cultivated), traditional farming areas—not surprisingly—retain the largest mean field size.624 The largest mean field sizes in acres were found in Blackwater (7.67), Casa Blanca (7.26) and
Sacaton (5.78). Santan, established in 1877, had a mean field size of 6.82 acres. Newly established villages such as Santa Cruz (3.21), Gila Crossing (3.74), Cooperative (4.75) and Maricopa (5.73) all had smaller mean field sizes in acres. Since the latter villages were of recent origin in 1914, it appears that, as water deprivation increased, villages not only moved downstream nearer the confluence of the Salt River but also were reestablished with smaller fields.