To accomplish anything like this within the Special Area would have meant a four fold increase in basic rations (calorie basis), or a doubling after the setbacks of 1939-40.
How, or even IF, this was accomplished musty remain a tricky matter for analysis, since, if Vladimirov is right, there was no central authority for provisioning the operational units (brigades, regiments, etc.). No doubt a lot of the food (just as with weapons) was scavenged from sources too far under the radar for the Central authority to monitor, even if it had wanted to. There seems in fact to have been a kind of mini-warlordism within the 8RA command, apparently at the Division level (5-7,000 mouths) emblemized by far-separated cantonments and minimal sharing of planning and of course ordnance: they seem, unlike warlords elsewhere, to have avoided armed conflict with each other, but cooperation was minimal, which explains why there were no more large-scale offensives after the 100 Regiments (!) campaign of 1940.
The three major-most units with toeholds in the SGN Special Area respectively the 115th “Division”(师), which was commanded by Lin Biao, the 120th under He Long, and the 129th under Liu Bocheng. According to an official source, these 3 “divisions” totalled no more than 9,000 regulars. But He Long’s division alone must have been several multiples of this (3,000/division) number. For he was indisputably deemed to have been the most powerful “feudatory”; a status achieved in part because of his early and sustained involvement in “secret-society” backed hill country partisan warfare in (Mao’s) Hunan. (See addendum).
(More complex is the role of the SA Political Commissar, Ren Bishi, who seems to have been the go-between with Mao. In 1934 Ren Bishi joined He in Guizhou with his own surviving forces after also being forced to abandon his soviet in another Encirclement Campaign. Ren and He merged forces, with He becoming the military commander and Ren becoming the commissar. in 1943 he was promoted to be the overall commander of Communist forces in Shanxi, Shaanxi, Gansu, Ningxia, and Inner Mongolia, meaning all of the Special Area with trimmings on the E. and N. . By the end of World War II He commanded a force of approximately 175,000 troops across northwestern China. He’s most notable subordinates included Zhang Zongxun, Xu Guangda, and Peng Shaohui. In October 1945, one month after the Japanese surrender, the command of He’s forces was transferred to Peng Dehuai, which operated as the “Northwest Field Army”. He became Peng’s second-in-command, but spent most of the rest of the Chinese Civil War in central Party headquarters, in and around Yan’an.)-(speculation: the Ansai yaoguwu was a kind of physical drill for army-recruits …? or at least an emblem of physical fitness.)
How large was the source loyal to or provisioned by He as the “120th”? The Nanniwan narratives speak of a total influx of non-indigene soldiers and party cadre and other “workers’ of about 100,000 in the first year or two of the Resistance War. Given that the SA population must have been 7-8 millions (today it is double that), and that the ratio of “non-producers” (non-peasant…) to total population ranged from the stipulated 3% to a peak of 6% in parts of the SA (Peattie, 319), and that the stipulated soldier/civilian ratio was 3/1 among party-payrolled “immigrants”, and that there were (below) approx. 1.4 million tax-paying (peasant) households in the SA and that the average number of “mouths” per household would have been about 5:
fn: Population size of the SA is anyone’s guess, since one never knows whether the reported numbers are based on fiscal survey or militia survey, or what. Zone sizes also vary tremendously, as (therefore) do estimates of population density.
1. http://zh.wikipedia.org/wiki/%E9%99%95%E7%94%98%E5%AE%81%E8%BE%B9%E5%8C%BA#.E5.86.9C.E4.B8.9A estimates the total SA population at 15 million, or about 115/km^2 over 130,000 km^2; the same number (129,000 km^2) is given by http://wuxizazhi.cnki.net/Search/NOSE201303007.html
2. http://xian.qq.com/a/20120319/000002.htm says 17 million in 192,000 km^2 or 90/km^2
3. http://baike.baidu.com/view/115898.htm says 92,500 sq km for ShaanBEI, the core of the SA, today’s population 5.62 million, or about 61/km^
For what it’s worth, the number of (grain) tax paying households is consistently reported as 1.4 million (1941-2)
details the end-1941 guidelines for troop strength reduction (or imposition of a ceiling): 3% of population for all public sector incl army; 2.25% for army (incl guerrillas)/ tot SA, then derived SA population = 3.3 million if supported forces = 100,000; (as in come cases) it went as high as 6%, then pop= 1.7 million.
Tax-roster enrolled is probably the basis of population calculation, and households (hu) the unit of measure, then likely population in all taxed households at 5/hh= 7.5 million.
Total APGC (non-productives):
(1) “Non-productives” should have been 0.025 * 7.5 or 150-180,000, some 3/4 of them (+/-120,000) soldiers, the rest – 40-50,000 – non-military cadre. That figure seems too high, since He Long’s 120th Div., the main unit, could not have been more than 15,000, including support personnel. In planning documents relating to the GIP campaign (Greatly Increase Production…) of 1941 ff, the number of “extraneous” (surplus) mouths on the supply roster is flatly stated as 100,000. I should take the lower figure, to allow for inflated supplies claims – a practice usual in all ramshackle armies.
(2) One almost accidental source of information about grain tax collection and supply for 1940 and 1941 – collected to justify the “Great Increase Production” drive of 1941- goes far in helping us excavate the guesstimates used by the Center. Converting the (unspecified) units of volume used for calculating in-kind taxes in millet, which formed the core of the 救国公粮 or “public food supply for National Salvation” tax fund, can be quite a nightmare, not the least because we have no way of defining which (and what pct) of landholding peasant households were on the tax roles, or of understanding how the enrollees got there. Reverse extrapolation gives the number of tax-paying peasants (=households) as 1.4 million, thus a total taxpaying population of around 7 millions, about half of today’s population – which seems right.
That being said, the numbers are still worth plumbing: they suggest a much higher per capita nutrition level for gongjia ren (government employees in toto, civilians and army troops) than one might expect, and (more speculatively) a drop in retained food/household of rather steep proportion.
Chart 1: Estimated Available Diet Calories (Millet), Yenan Special Area; Public vs Peasant Sectors