96. H. B. W. Joseph to Harrod , 27 November 1925 [a]
[Replies to letter not found, follows on from 95 ]
I'm sorry no day suggested is any use. What about Thurs. or Friday next week?
I don't think that if I find my preference of a against b about the same as my preference of b against c, my preference of a against b is twice as great as either. [1] I may think, on a scale of colours differing from lighter to deep blue, that the difference between a & b is about the same as that between b & c--I'm not quite sure what that means, but perhaps that there are as many just distinguishably different shades between in each case. But I can't say that a is twice as much darker than c as it is than b: only that the intervening shades are twice as many + 1 between a & c as between a and b. Equally distinguishable differences are not, for all I know, equal: & if they could be said to be, I can add them in number, but not in quantity.
I am very unsure what the total detriment of the people means. There is no aggregate happiness of a community. There is an aggregate of persons, each more or less happy. But there is no happiness not experienced, & no one experiences their aggregate happiness. I think I can say that a community whose members are all happy is a better thing than one whose members are not. But I don't reach this judgment by adding happiness. Its goodness is something one, not an aggregate. So it seems to me better that the burden of taxation should be so distributed as not to make the poor poorer & the richer relatively richer. But I don't think a government could reach its best distribution of the exaction of wheat by a sum. Nor if the citizens differ in wealth, health, temper & responsibilities, do I know how it could determine what relative exactions would inflict equal detriment on all, if detriment is not measured in terms of wheat taken, or meals omitted, but of "sacrifice".
However, it would really be better to proceed verbatim.