Content area
Full Text
In the first issue of 1999 of this journal, Thomas Whigham and Barbara Potthast published a research note on the demographics of the Paraguayan War (1864-1870). They reported a newly discovered census from 1870, presented an analysis of the census results, and concluded that the old stories of a steep loss of population during the war are basically correct. The census data showed that Paraguay's total population in 1870-1872 had declined to 116,351 persons (see table 1). Because several villages had failed to report to the Asuncion authorities, Whigham and Potthast adjusted the census. They assumed that 25,000 to 50,000 Paraguayans remained in these areas, making the total population somewhere between 141,351 and 166,351. They concluded that even after making these additions, the loss would seem to be 60 to 69 percent of the Paraguayan population before the war.
When I started reading this piece, I was glad that new census materials had been discovered and that new insights had become available because the new information would help me in revising a section of my forthcoming work entitled "Paraguay, 1515-1870: A Thematic Geography of Its Development."1 But after reading the entire research note, I had major reservations and felt somewhat disappointed.
Whigham and Potthast have invested much energy in correcting the 1870 census data by adding missing figures, but they hardly raised the question of whether the available data are sufficiently reliable. In my opinion, they overestimate the reliability of their figures. My reservations are based on three arguments.
First, all "censuses" taken in Paraguay in the eighteenth and nineteenth centuries were characterized by undercounting. Whigham and Potthast are aware of this fact, which they illustrate in mentioning several weak points of the 1846 census, including the undercounting of children. To cite another example, Jose Jacquet, the statistician responsible for the census of 1886, knew that a lot of people had been omitted, certainly in the inaccessible areas of the countryside. He therefore immediately increased the 1886 census figure of 239,774 inhabitants by 10 percent. In 1887, however, he decided that this correction was still too low, and he increased the original 1886 figures by 37.5 percent. Jacquet's second correction brought the population up to 329,645 in 1887 (excluding Indians in the north and west...