What's going on with this graph of wheat prices in "Civilization and Capitalism"?

score:4

Accepted answer

Ah, I think I've figured out the error here - the key turns out to be the spacing of the horizontal lines. (I'll leave this question up, in case it serves to clarify things for any future readers who investigate this issue.)

As is apparent from the spacing of the lines, the y-axis is intended to be logarithmic. If we take the markers 2 though 9 to be accurate, then we can obtain consistent measurements for this scale: on my copy, I measure a factor of 2 (such as the distance between the 2 and 4 marks, or the 3 and 6, or the 5 and 0) to be around 2.2cm, and a factor of 3 to be around 3.5cm (as we should expect, the ratio between these lengths is very close to the quotient log(3)/log(2)).

If we then extrapolate these ratios above the 0 line, we find the following distances:

0  to 10: 2.2cm
0  to 20: 3.5cm
0  to 30: 4.4cm = 2 * 2.2cm
10 to 30: 2.2cm
10 to 50: 3.5cm
 5 to 20: 2.2cm
5* to  5: 3.5cm

*Referring to the lower of the two marks labeled "5".

These measurements are all consistent with the following relabelings (maintaining units of tens of hours of manual labor):

 "0" -> 10
 "5" -> 15
"10" -> 20
"20" -> 30
"30" -> 40
"40" -> 50
"50" -> 60

If we make the above relabeling, everything makes sense: the scale is nicely logarithmic, the crossings occur at 100 hours, the peak occurs near 500 hours (in fact it goes slightly over the threshold, rather than slightly under as it appeared originally), etc. I would conjecture that the original graph was accurate, and in the process of typesetting, some labels were mistranscribed, or perhaps that the labels above the "0" line were meant to indicate the excess above 100 hours, and this change was not propagated to the original description.

More post

Search Posts

Related post