Peter Hart. The I.R.A. at War, 1916-1923. Oxford: Oxford University Press, 2003. xvi + 274 pp. $40.00, cloth, ISBN 978-0-19-925258-9.

Reviewed by William Kautt

Published on H-War (March, 2005)

For almost seven years, between 1916 and methodology, on the war. He does this in the 1923, Ireland endured major changes in its politi‐ hopes of sparking a new debate about the nature cal and social fabric. These changes were brought of the confict. Indeed, in the title of his introduc‐ about primarily through violence and, to a certain tory chapter he calls this "A New Revolutionary extent, are still felt today. This upheaval History." Here, Hart explains that Ireland's fght wrenched most of Ireland from the United King‐ for independence is important in a wider context dom and inexorably changed its political land‐ because, in a very real way, the I.R.A. "invented scape. In the last thirty-fve years, there has been modern revolutionary warfare" (p. 3). He contin‐ a re-examination of this confict with a view to‐ ues, arguing further that its relevance is frmly ward ridding the feld of politically favored ac‐ rooted in the fact that it was one of the best-docu‐ counts. Most recently, a new generation of histori‐ mented wars of its kind in history. ans has taken the lead in this re-evaluation. Fore‐ However, before proceeding, it is important most among them is Peter Hart, who burst onto to note that this work is not military, but rather, the scene with his masterful work, The I.R.A. and social history. In some respects, this hurts his its Enemies (1998), on the social makeup of the analysis when he delves into more military topics. I.R.A. confict in . Since the publica‐ In fact, Hart, in questioning the usefulness of tion of his frst text, Hart has become the stan‐ guerrilla theory, appeals to a social scientifc ap‐ dard-bearer for the empirical study of the confict proach. Importantly, he says, "if nationalist expla‐ and has made major contributions in the feld. nations for the course of events fail to meet em‐ Almost anticlimactically, Hart's latest book is pirical and logical tests, they should be discarded" an anthology of his previously published articles, (p. 5). Yet what he does not recognize is that by di‐ albeit with critical updates and additions. The ma‐ vorcing theory--partly the explanation of actions-- jor strength of this compilation is that it brings to‐ from guerrilla warfare, one loses the meaning gether his best ideas, mostly using statistical and intent of sometimes obscure or seemingly in‐ H-Net Reviews defensible actions by the insurgents. Still more bias into his results. In his newly revised chapter importantly, the symbolic nature of actions is lost. 2, Hart corrects this defciency by examining all With this slightly narrow-minded view, Hart then thirty-two counties in Ireland. In the process, he proceeds to describe his vision of revolution, confrms his earlier fndings. which actually difers little from commonly ac‐ Unfortunately, Hart's methodology is still not cepted defnitions. completely sound, because he only examines 1917 Despite this shortcoming, Hart's introduction to 1923, thus ignoring the critical 1916 Easter Ris‐ rightly points out that the war has no proper ing, despite acknowledging the Rising as the be‐ name--it is variously known as the "Anglo-Irish ginning of the revolution. Hart also errs by using War" (which is unacceptable because it implies problematic sources such as the monthly police that it was a war solely against the English); the reports from 1917 to 1921 (they ceased in 1922); "War of Independence" (objectionable because it the Irish Times, the Cork Examiner, and the implies that the republicans won the war); and Times, all from 1917 to 1923; and the Freeman's the "Tan War" (which is too specifc, being the Journal for 1921-23. These sources not only cover name of only one group of combatants) (p. 4). diferent times, but also are disparately biased. Such confusion and contradiction in terms per‐ For example, using only one regional newspaper fectly and appropriately captures the perplexing (the Cork Examiner), which covered only some of and often controversial nature of the confict. the six counties of the province of Munster cannot Hart makes the case for using "Irish Revolution" possibly capture what Hart purports to be mea‐ because of the tremendous change in Irish society suring. He should either have eliminated that and politics resulting from the confict. Perhaps it source or found its equivalent in the three other is reasonable to use this general term because provinces. Thus, his data are not a representative there were multiple phases of the confict--specif‐ sample and, as a result, he reintroduces sample cally the 1916 , the 1919-1921 "Tan selection bias into the study. From this also stems War," and the (1922-23) to name a concerns about content validity. Clearly, his claim few. of 90 percent accuracy is impossible to substanti‐ One of Hart's greatest contributions to Irish ate empirically or otherwise. historiography has been his judicious use of statis‐ That said, this does not mean that Hart's con‐ tical analysis. The second chapter, "Geography of clusions are wrong. Based on the available quali‐ Revolution," is a revision and expansion of his tative evidence (of which there is a great quantity, May 1998 Past and Present article of the same hence the need for statistical means to examine name. The original article was hailed as ground‐ it), his original article and his new evidence are breaking in its use of statistics to track and map likely correct. Therefore, one cannot, nor should violence in Ireland during the war. It confrmed one attempt, to dismiss the study. Rather, Hart de‐ what was popularly known from the time of the serves great credit for introducing statistical war, that violence and its manifestations were methods into recent Irish historiography. quite varied by location throughout Ireland in Hart's general weakness in military history terms of number and frequency. There was, how‐ resurfaces in chapter 3 with a faw that touches ever, a faw in the original statistical methodolo‐ on a question about when the middle confict gy; Hart chose his sample of locations for deliber‐ (generally 1919-1921) actually began. Hart states ate, and therefore biased, reasons. By doing this, that by measuring levels of violence, one could ar‐ rather than selecting randomly or examining the gue that the war really began only in January entire population, he introduced sample selection 1920, rather than January 1919. Here he makes

2 H-Net Reviews the error of equating a low level of violence with perhaps, minor points, but are indicative and are an absence of war. Simply put, causing a stone actually made more apparent by Hart's outstand‐ bridge to collapse by removing the keystone is not ing prose and otherwise fawless logic. directly violent, but if it blocks an avenue of ap‐ A less problematic issue, which is also refec‐ proach and secures the I.R.A.'s fank in an arms tive of this general weakness in Hart's work, is raid on a police barracks, it is no less ofensive that he only counts those wounded or killed by than blowing it up; or indeed leaving a blocking bombs and frearms as casualties of the confict. force to attack a relief party. Hart misses the con‐ Yet there were several well-documented instances cept of preemptive or preventative action in guer‐ of I.R.A. members and police being burned when rilla war, which is a critical omission because the latter used arson against R.I.C. barracks dur‐ much in this type of warfare is preparatory or ing raids. Were these injuries not wounds re‐ symbolic. His later description of guerrilla war as ceived in action? There are many ways to wound "mass homicide" (p. 89)--referring to the numer‐ or kill, but Hart's justifcation that gunshot ous assassinations and executions--only serves to wounds and bombings are almost always report‐ amplify this point. Obviously, one need not be a ed is well taken, if not entirely correct. It is impor‐ military historian to write about war, but clearly tant not to overstate the case, but methodology is one should demonstrate a sound knowledge and still important. understanding of it when addressing a military Another statistically based chapter on I.R.A. topic. violence follows and compliments the previous Although Hart's misunderstanding of the na‐ geographical one. Likewise, Hart's chapter 5 on ture of military strategy in general and guerrilla Irish republicans and their activities in Britain is warfare specifcally is present throughout his a very good survey of the subject, but adds little to work, it only hampers analysis in specifc cases. his earlier article of the same name ("Operations For instance, by questioning the war's starting Abroad," English Historical Review, 115 (2000): date, one would have to accept that the British pp. 71-102). Cabinet, in late 1919, radically altered long-stand‐ However, one chapter that stands out is his ing policy by reinforcing the greatly beleaguered demographic survey of the IRA rank and fle. By Royal Irish Constabulary (R.I.C.) with non-Irish‐ charting their social background, including eco‐ men without a sufcient reason. Yet even these nomic activities and employment, familial con‐ reinforcements, arriving in early spring 1920, nections, and religion, Hart gives the reader an were insufcient to stem the tide of the on-going excellent view of these men. Indeed, a general R.I.C. general withdrawal (they themselves some‐ work on this topic is much needed. He goes into times referred to it as a "retreat") from the coun‐ greater detail than his original and excellent ex‐ tryside and smaller towns to cities, which began position in his I.R.A. and its Enemies, which con‐ in January 1920. Thus, his claim defes logic, for tained a great deal of such information, but was why would the police retreat unless they were al‐ limited to County Cork. ready losing forces in the war? It is highly unlike‐ ly that the R.I.C., experiencing less than a month At frst, chapter 8, on the long-standing ques‐ at war (going by a January 1920 start date) were tion of whether Michael Collins ordered the mur‐ so demoralized as to require a general withdraw‐ der of Northern Irish loyalist Field Marshal Sir al. Further, the Inspector-General of the R.I.C., Henry Wilson in 1922 seems out of place because Brigadier-General Byrne, was so depressed by the it reads more like "true crime" than history. But confict that he was relieved in March 1920 after his excellently written discussion, almost in the having held the position for four years. These are, form of a crime report, takes the reader through

3 H-Net Reviews the competing theories, and in so doing, bridges and northern Catholics. Also, I accept his explana‐ the gap between the end of the I.R.A. war against tion of the geography of violence with the provi‐ British rule and the beginning of the Irish Civil sos already stated. Nor can I fnd fault with his War in 1922. chapter on the "operations abroad," except to say This chapter on the murder of this champion that much remains to be done in that area, but his of the loyalist north leads naturally to his last two introduction of the topic is very good. Finally, I chapters on the experience of Protestants, who agree with his argument in choosing "Irish Revo‐ were mostly loyalist, in the heavily republican lution" as the appropriate name for the confict. In south. In detailing this, he brings the ugly past of the end, I see little to debate because Hart has ba‐ republican-initiated violence to the fore. Although sically gotten it right. not ofcially sanctioned by the provisional gov‐ Aside from the issues about the nature of ernment, this is a topic that most researchers guerrilla war and the sampling methodology, this have avoided. Hart addresses it head-on and is an excellent work. And although not a military brings his impressive statistical skills soundly to history, nor really meant to be, it is absolutely es‐ bear once again, demonstrating the magnitude of sential for any serious student of the war or of the the situation and juxtaposing it briefy with the era in Ireland. plight of Catholics in the north. In the fnal chapter, Hart examines these northern Catholics and their response to violence, privation, and harassment. He places both the southern Protestant and the northern Catholic persecutions into context by comparing them. His treatment of these two related topics is impres‐ sive and completely objective--a rare quality on such a sensitive issue. This is, no doubt, due to his statistical basis for the evidence. Hart's book sufers somewhat from the dis‐ jointedness of seemingly disparate topics that, in reality, should be joined together by the common thread of the Irish Revolution. Several chapters are repetitive, an issue that would not be present if written or edited in a more complementary fashion. Adding to this problem is the complete lack of a concluding chapter, which causes an un‐ timely end that squanders the opportunity to unite the various factors. Hart wanted to start a new discussion of the era of the revolution as a whole. Instead, he sim‐ ply brings some of his best work together into one volume. Whether his work will spark debate de‐ pends largely on how his results are received. I cannot fault his demographic work, nor his dis‐ cussion and treatment of the southern Protestants

4 H-Net Reviews

If there is additional discussion of this review, you may access it through the network, at https://networks.h-net.org/h-war

Citation: William Kautt. Review of Hart, Peter. The I.R.A. at War, 1916-1923. H-War, H-Net Reviews. March, 2005.

URL: https://www.h-net.org/reviews/showrev.php?id=10377

This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License.

5