Chinese Warlords: Tigers or Pussycats?

CHINESE WARLORDS: TIGERS OR PUSSYCATS? by James E. Sheridan The editor has invited comment on Rawski's essay in the hope of clarifying the nature of warlordism. Rawski argues in favor of five hypotheses that relate to major features of the entire republican era, not warlordism alone. I will not attempt to discuss all those hypotheses. Here I wish only to consider a specific aspect of Rawski's essay that emerges particularly in his discussion of his second and third hypotheses, where he essentially argues that warlords have received a bad press from a number of scholars, primarily historians. Rawski claims that warlord activities have not been as harsh and explo~tative as often described, and the effects of warlord rule have not been as negative as most writers have claimed. During the republican period, Rawski argues, there was no major departure from the "venerable Chinese tradition of low taxes" [p. 28]. He maintains that a number of warlord studies show "unfortunate anti-~ilitarist bias" [p. 51], overstate the "negative features" of military regimes and neglect or dismiss their "progressive aspects" [p. 55]. It is this--Rawski's rather benevolent view of the warlords (and his rather harsh view of warlord studies)--that I want to discuss. I find his argument unpersuasive for several reasons. The first thing to note is that Rawski and the authors that he criticizes have different interests; they ask different questions, and use different kinds of evidence. Rawski is concerned with "resource flows," "total output and income," and the like. One of Rawski's approaches is to generate various total figures--total output, total government spending, etc.--compare them, and draw inferences from the comparisons. For example, his figures for total central governm~nt revenues and spending are small compared to gross domest~c product. On the basis of this, and a similar analysis of provincial revenues and outlays, Rawski concludes that "taxes generally took only a small share of output" [p. 35]. Thus, presumably, they could not have been very oppressive. In a similar fashion,. Rawski estimates total troop numbers and geographic areas of warfare, and concludes that they were small compared to the population and area of China, and therefore could hardly have been as exploitative and damaging as historians have claimed. The warlord writers, by contrast, are concerned about people, about the way in which warlord activities affected the daily lives of millions of Chinese of all classes and situations. They use not only some numerical data, but reports from diplomats, missionaries, journalists, travelers and others; diaries and reminiscences, including those of warlords themselves; newspaper and periodical accounts, and a myriad other sources. The two concerna--"resource flows" and impact on people's lives--are obviously related, and one 35 might infer much about one from information about the other; that is precisely what Rawski has done. But the two approaches are baaed on different kinds of evidence. Which is the more believable? When one wants to know the effect of warlord sense to accept the testimony participants, and various kinds of Rawski criticizes have done, or acknowledgedly crude and inadequate rule on people, does it make more of a host of witnesses and primary evidence, as the scholars to accept inferences drawn from data, as Rawski does? This is no place to wage a battle of quotations; it makes no sense to repeat the diverse evidence that has been offered by scholar after scholar about the harsh effects of warlordism. But one cannot help but ask why, to take a single example, one should reject the testimony of the co-Director General of the Chinese Post Office, whose professional activities presumably gave him knowledge of conditions in many parts of the country, when he says that in 1917-not an atypical warlord year--there was a "recrudescence of highway robbery all over the province" of Shantung; that in Shensi "bands of t'u-fei [local brigands] roamed through the province, plundering and looting"; that in Szechwan "there was scarcely a place unaffected by military operations"; that Hunan province was "from the beginning of the year •.• in a state of unrest and under martial law"; that northern Anhwei was "in a state...