Summary

In summarizing the discussion, it is convenient to divide the errors exposed into those that refute the whole of Sen's thesis and those that refute only part of it. Each of the following errors is, by itself, fatal to Sen's whole thesis. If any one of them is accepted then his whole thesis must be rejected. It is not, of course, possible to argue in economics that his theory may be right but his facts are wrong.

• Production statistics are not accurate to within +/- 50%, and the difference between them (which Sen relies on) is only accurate to +/- 3000%.
• Sen's assumption of zero carryover conflicts with all available evidence.
• Sen's production and import figures do not prove his point that there was food available in 1943, but exactly the opposite.
• Apart from the unreliable production statistics, all evidence (including that on speculation and inflation) points to the fact that a) there was a short crop and b) there was a shortage.
• If changed distribution caused the famine, some groups of the population ate between two and six times as much as usual and paid between four and 20 times as much as usual to do so. It can be proved both logically and statistically that they did not.
• The actions of the government of Bengal were those Sen would recommend. Their failure to have any effect proves the misdiagnosis, and my prediction of the effect of the misdiagnosis.
• There are repeated misstatements and misquotations from his sources - on the indifferent crop', on his conservative figures, on the number of people covered by relief schemes, on the actions of the government of Bengal, on the Famine Commission's support for his statements about speculation and hoarding, on the rice denial policy, on the size of other famines and on Mahalanobis, Mukkerjee and Ghosh's statistics for instance. In addition, the evidence presented is selective. Taken together they cast the gravest doubt on his rigour and reliability.

The following points taken individually do not disprove the theory as a whole, but only individual hypotheses:

• On inflation: Sen presents no theoretical explanation of an improbable hypothesis; the evidence he does present is incorrect; and other evidence indicates a shortage. Government procurement plus a shortage is the best explanation of the enormous price rises.
• On speculation: Sen presents no theoretical explanation; his thesis conflicts with accepted theory; and his thesis conflicts with the evidence. The evidence plus accepted theory suggests a shortage.
• On hoarding: Sen presents no theoretical model and the facts are against him.
• Changes in purchasing power are more likely to be the effect rather than the cause of the famine.
• Imports were unnecessary under Sen's thesis.
• The boat denial policy did not reduce aggregate supply of rice. Failures in regional distribution are irrelevant to Sen's thesis.
• Sen's hypothesis and facts on the rice denial policy are contradicted by his sources. If anything, the policy increased total supply.

It has not been the aim of this paper to appraise Sen's entitlement theory. It should be noted, however, that even in the hands of its originator it is apparently incapable of detecting the many contradictions in the model presented or the many factual errors. Modern marketing theory, on the other hand, exposes the errors and contradictions immediately.