EUROPE-ASIA STUDIES , Vol. 51, No. 2, 1999, 315±345
Victims of Stalinism and the Soviet Secret Police: The Comparability and Reliability of the Archival DataÐNot the Last Word STEPHEN G. WHEATCROFT
ONQUEST’S T’S COMMENT ON MY LAST LAST ARTIC ARTICLE LE in thi C ONQUES this jou ourn rnal al 1 raise raisess so some me important important questions for our profession. Are we going to progress in our level of understanding? Are we going to respond positively to the new circumstances in which large amounts of detailed archival materials are available? Are we going to try to critically assess th e r el iabi ia bili lity ty of the th e se data da ta?? A r e w e goin go in g to try tr y to pr ov id e cr e dibl di bl e indi in di cato ca to r s of the th e Soviet experience that we can compare with other societies? Conquest’s response to these questions is disappointing, but not totally unexpected. In principle, he agrees that older work `must be subject to major amendment’ as new material becomes available. available. But in practice practice he is very resistant resistant to changing changing his earlier earlier views, despite the overwhelming evidence that his earlier estimates, and his minor `reassessments’, are still far too high. His comment on my article suggests that he is no t r ea lly inte in te re sted st ed in the th e que qu e stio st ions ns r aise ai se d. He doe do e s no t se r io usly us ly disc di sc uss us s the th e reliability of the data, with which he is clearly unfamiliar, and which he wishes to dismiss out of hand. In his brief Comment, Conquest claims a lot. He claims that the argument of my arti articl clee is `fu `fundamen ndamenta tall lly y ¯awed’ ¯awed’.. He clai claims ms that hat I `cl `claim aim to pres presen entt the the true, rue, ªarchivalº totals for the victims of Stalinism’ and that this claim is `fallacious’. He claims that my sources can be reduced to three documents, which are incompatible, incomplete and consciously faked. And he wishes to replace them by a set of other les less `det `detailed ailed’’ and and less less `preci `precisse’ ®gures ®gures,, which which he think hinkss are more credi credibl ble, e, and and which, incidentally, support his earlier `higher’ estimates of the scale of the camps and of mortality in the repressions. In a somewhat contradictory manner he admits th at ` th e es tim a tes te s I [ Co nq u e st] st ] ar ri ved ve d at on K oly ol y m a w e re inde in de ed ex ce ssive ssi ve,, an d as with other early estimates on the whole terror period, now that more is known they must indeed be subjected to major amendment’. But he then goes on to claim that `in every every othe otherr way my book remai rem ains ns a full full and and now no w full fully y veri®ed eri®ed acco accoun untt of the the subject’. His comment then goes on to attack me for what he considers to be my `conceptual error ¼ on the system’s casualties over the USSR as a whole’. And in conclusion conclusion he adds that he also also doesn’t doesn’t like m y continual refusal refusal to accept that that `Stalin `Stalin consciously in¯icted the 1933 famine’, to which he addresses a few more comments. Conquest’s comments do not particularly disturb me. In fact, for those who can read the nuances, it will be apparent that his criticism is relatively mild. After all, 17
0966-8136/99/020315-31 $7.00 Ó 1999 University of Glasgow
31 6
STEPHEN G. WHEATCROFT WHEATCRO FT
years ago, when, as a young research student, I ®rst publicly addressed his misconceptions over the scale of forced labour, he was far more outspoken. At that time he accuse accused d me of `disto `distorti rting’ ng’ and `inadeq `inadequat uately ely checki checking’ ng’ stati statisstical tical evidenc evidence, e, and of using arguments which `amount to no more than a combination of sophistical algebra and unwarranted parochial assumptions’. 2 Thi T hiss w as in r e spon sp onss e to m y a r gu m e nt th at hi s ® gur gu r e of ei ght gh t m illi il lion on in the th e la bou bo u r cam ca m ps ( nin ni n e m il lion li on if yo u in clud cl ud e pr ison is on s) was incompatible with the arguments about the scale of the camps which had been made earlier by Jasny, Timashev and Eason, and with an assortment of other evidence th at I liste li ste d. O n th e ba sis si s of a c ar eful ef ul an al ysis ys is of th e na tu r e of S ovi ov i e t s ocie oc iety ty an d the th e economy of the 1930s I concluded that it was impossible for there to have been more th an f ou r to ® ve m illi il li on in th e la bo ur cam ca m ps in the th e late la te 1930 19 30ss . 3 Thi T hiss is a co nc lusi lu sio o n th at ha s no w bee be e n tota to ta lly vi ndic nd ic at ed, ed , an d tha th a t ev en Co nque nq ue st oc c asio as io nall na lly y ac ce pts. pts . I did not expect Conquest to make an apology, or to acknowledge his error, but his repeat repeateed ins nsiistenc stencee that that the curre current nt ®gures ®gures have sh shown own him him to be correc correctt on this his question is a little hard to swallow. Fr F r o m his hi s r ece ec e nt c om m en ts it is dif di f ® cu lt to un pic pi c k w ha t he no w thin th in ks is m y `conceptual error’. He is clearly annoyed that I continue to challenge his ®gures, and in desperation has moved on to attack me for things that I have not said. Conquest’s stat stateme ement nt that that I `cla `claiim to presen resentt the the true, rue, ªarc ªa rchi hiv valº alº total otalss for for the vict victims ims of Stalinism’ is ridiculous, as will be shown below. From his comment and the whole th ru st of hi s r ece ec e nt w r it ings in gs,, it a ppe pp e a rs th at C onqu on qu es t is still st ill cl a im ing in g th at a ltho lt houg ug h hi s K olym ol ym a ® gu r es ar e w r ong on g , th e r est of his hi s ear ea r lier li er esti es tim m at es a s re s tated ta ted in The Great Terror: Terror: A Reassessment Reassessment (1990) are correct. If this were all, it would not matter so much, and we could leave Conquest to his dreams, but unfortunately other in¯uential scholars appear to be accepting Conquest’s claims that the new data con®rm his `high ®gures’.4 And so I feel obliged to put the record straight (again). My response to Conquest is long, because most readers of this academic journal will ®nd it dif®cult to make sense of his brief comment. They will come away from it w ith ith the sens sensee that that `the `the big biggest gest name name in the profes professsion’ ion’ thi thinks that hat the work of Wheatcroft Wheatcroft and others others w ho atte attempt to analy analyse se the the archiv archival al data data is `fundamen `fundamental tally ly ¯awed’ ¯awed’ and and su suffe ffers rs from from `conc `concep eptu tual al errors errors’. ’. It will will not matte matter to them them that hat the the te ch nica ni call ar gum gu m e nts nt s see m so co m ple pl e x that th at they th ey ca nno nn o t f ol low lo w them th em . T h e ha r m w ill ha v e be en done do ne Ð Co nque nq ue st w ill ha ve s how ho w n that th at he can ca n still st ill an s we r his hi s cr itics it ics , an d th at his hi s ear ea r lier li er as sessm se ssm en ts or ` re asse as se s sm smen ents ts’’ ar e co rr ect. ec t. I hop ho p e that th at the th e m or e th ough ou gh tful tf ul of th e r ea de rs w ill il l go be yon yo n d this th is an d w ill at tem te m pt to un de rs tand ta nd the th e arguments about the value of these new sources. In this article I argue that despite all the smoke-screens Conquest is continuing to adhere to his old `high ®gures’ or his 1990 `re-assessed high ®gures’, which are both no w un te na ble. bl e. A t va r ious io us tim ti m e s, in var va r iou io u s pl ac es Conq Co nque ue st ha s m a de sta st a tem en ts th at approa approach ch a more realist realistiic apprec appreciat iatio ion n of the situati situation, on, but he then then quali quali®es ®es thes these statements and neutralises them by recon®rming his belief in his earlier ®gures. This leads leads him him to reject reject a whole body of archiva archivall informati information on in favour favour of hearsa hearsay y and unsubstantiated unsubstantiated literary literary sources. Conq Conquest uest is wrong wrong in stat statin ing g that that I am clai claiming ming to prese presen nt `the `the ult ultimate imate trut truth’. h’. Conqu Conquest est is wrong in clai claimin ming g that that the new new evid eviden ence ce vin vindicat dicatees his earli earlieer `hig `high h series series’’ of camp camp and and mortali mortality data. Conques Conquestt is w rong in stati stating ng that that m y articl articlee is
VI CT IM S O F TH E SO V IE T SE CR E T PO L IC E
317
based simply on accepting Kruglov’s report (1954), the Shvernik report (1963) and th e Z em sk ov ® gu re s ( 1989 19 89 , 19 9 1 an d la te r ). He is w r ong on g in su gges gg es ting ti ng th at th e da ta in thes thesee repor reportts can can be easil easily y sh shown own to be false false,, and and so sho hou uld be dismis dismisse sed d as fabricated and of no use. He is right in stating that there is a partial incompatibility between these data, but he is wrong in his statements concerning the scale and size of this this incompat incompatibi ibilit lity, y, and what it implies implies for the the Gulag Gulag data. data. 5 And ®nally, on a different topic, which Conquest has nevertheless raised here, Conquest is wrong in rejecting out of hand those who provide evidence that Stalin did not consciously plan th e 193 19 3 3 f am in e. All these errors need to be addressed, but I will try to do this in a positive way, which will give the reader a greater understanding of the complexities of some of the important problems that are being discussed. I begin with a brief discussion of the hi stor st or y of w r itin it ing g s on th e ` ca sua lty lt y ® gu r es ’ an d of Conq Co nque ue st’s st ’s cl aim ai m s c once on ce r nin ni n g th ese es e ® gu r es. es . F o r ne w co m ers er s th is ca n be seen se en a s the th e plo pl o t so f ar . I the th e n c ons on s ide id e r the th e na tu re an d th e or igin ig in s of the th e a rc hiva hi vall sour so ur ces ce s , in pa r ti cula cu larr thei th eirr re lia bili bi lity ty an d compara comparabi billity, ty, and and the the charg charges es made by Conqu Conquest est,, that that they hey can can be dis dismisse missed as incompat incompatib ible, le, incomplete incomplete and fake. fake. Next I consi consider der the the alte alternative rnative sources sources which Conquest claims are superior to the archival sources. The following section deals with a totally different question, mainly the discussion over control and intention regarding th e f am in e. F inal in al ly I r e- s tate tat e the th e m a in c oncl on cl usio us io n that th at I be lie li e ve sh oul ou l d ¯ ow f ro m m y anal analys ysis is of the dat data and and compar comparee it with with some of the the po pop pular ular misun misundersta derstand ndiings regar regardi ding ng i t , that that are stil stilll being eing repe repeat ated ed.. The appe appen ndice dicess cont contai ain n so some me of the the statistical data, as well as a consideration of some of the wilder conclusions that are curren currentl tly y bein being g made, made, by othe otherwis rwisee san sane acad academi emics cs,, as a resu result of the clai claims ms of Conquest.
A brief comment on the history of writings on the scale of the Soviet labour camps and Stalinist repression
Conquest’s major volume on The Great Terror has been treated as a classic ever since its its appe appear aran ance ce in 19 1968. 68. His appe append ndix ix on `Casua `Casuallty Figure Figures’ s’ has has been been enormo enormous usly ly in¯uential and captured the imagination of many readers. Let me say at the outset that in the past I found this work useful. It contains a good account of most of the political and so socia ciall litera literatur turee which which was avail availabl ablee at the time. U nlike nlike many other other poli politic tical al works of the time, it paid particular attention to the scale of the terror. 6 Given the poor availability of material then, it was quite an achievement to produce such a survey. One of my earliest tasks as a research associate working for R. W. Davies and Moshe Lewin about two decades ago was to check on Conquest’s casualty ®gures. At the tim ti m e I co nc lude lu de d th at th e of ® ci al S ovie ov ie t vi ew w a s w r ong on g a nd that th at Co nque nq ue s t w as correct to argue that the scale of violence was of demographic signi®cance, but I also conc conclu luded ded that hat his eval evalua uati tio on of that that scal scale was inacc inaccur urat ate. e. 7 Subsequent Subsequent archival archival materials have shown that Conquest’s estimates did indeed exaggerate the size of the labour camps and mortality in the labour camps, in the manner in which I had earlier expl explai ain n ed. ed. Howeve Howe ver, r, it must must be sai said t hat hat Conq Conques uestt’s est estimate mate of the the numbe num berr of executions in 1937±38 was somewhat less exaggerated than I and others had earlier expected.
31 8
STEPHEN G. WHEATCROFT WHEATCRO FT
Conquest’s ®gures for the labour camps rested partly on those given by Professor Ta T a lgr lg r en ( S wi a niew ni ew icz ic z ) an d othe ot he r s in th e 194 19 4 9 c la ssic ss ic by D al lin a nd N icol ic ol aevs ae vs ky , 8 T he Da llin ll in an d N ic ola ol a ev sky sk y w or k ha d esti es tim m a ted te d Forced Forced Labou Labou r in the Soviet Soviet Union. The a level of 10 million in the labour camps in 1940. However, since the appearance of th is cl as sic th er e ha d be en m uc h ser se r iou io u s w or k on the th e na tur tu r e of the th e S ov iet ie t e co no m y and society which had greatly improved our general understanding of how this society operated. Although many of these early revisionists were ®ercely hostile to Stalin’s Rus R ussi siaa they th ey ne ve r the th e less le ss re fu s ed to ac ce pt the th e po pula pu la r pict pi ct ur e pr om oted ot ed by D al lin an d Nicolaevsky, and insisted that there was little space for more than 4 million labour camp inmates in 1940. In 1948 1948 the the so soci ciol olo ogist gist Nich Nichol olas as Timas Timashe hefff us used ed data data on Sovi Soviet et elec electi tion onss to calculate a scale of disenfranchised population that only allowed space for 2 million 9 in the labour camps. The T he e co no m ist is t Jasn Ja sny y w as gi ve n a co p y of th e c ap tur tu r ed secr se cr et 1941 1941 Plan Pl an of t he Sovi Soviet et Natio Nationa nall Economy Economy,, whic which h cont contai aine ned d Gulag Gulag prod product uctio ion n tar ta r gets ge ts,, by hi s f or m er f r ie nd D al lin, li n, w ho f ul ly ex pe cted ct ed Ja s ny to co n ® r m his hi s ® gu r e of 10 million. However, the courageous Jasny increased his unpopularity in the USA by insisting in 1952 that it was impossible for this plan to cover more than 3.5 million 10 11 forced labourers. To T o c la im suc su c h ` low lo w ’ ® gur gu r e s w as ve ry unpo un po pu lar la r a t the th e tim ti m e, an d especially after 1956, when Khrushchev himself had admitted the nature of Stalin’s ho rr or s . T he e cono co no m ist is t B er gson gs on a nd the th e de m ogra og ra ph er E a son so n , w ho w ere er e w or ki ng f or th e R A N D C or por po r a tion ti on ( U S A ir f or ce ), w ere er e unde un de r s ta ndab nd ab ly a li ttle tt le he sita si tant nt a bou bo u t giving too much publicity to this aspect of their ®nding, in the late 1950s and early 12 1960s. In 19 1965 65 Swiani Swianiewi ewicz, cz, who had had earli earlieer (und (under the the pen-name en-name of Talgre Talgren n) pr ov id e d m a ny of the th e stat st at istica is tica l es tim ate s f or D a llin ll in a nd N icol ic olaa evs ev s ky , a cc ep ted te d m uc h of the reasoning of Jasny and revised his earlier estimates of 10 million down to 6.9 million. 13 Conquest in his survey in 1965 largely ignored the more serious (and complex) work of Timasheff, Jasny, Bergson, Eason and even Swianiewicz and presented what he r ep ea tedl te dly y de scri sc ri bed be d as a ` co nse ns e r va tive ti ve’’ eva ev a lua lu a ti on of eigh ei gh t m il lion li on in the th e cam ca m ps in early 1939. He also claimed a level of seven million arrests between January 1937 and December 1938, one million executions and three million deaths in the camps in 14 th is pe ri od. od . The T he labo la bo u r cam ca m p e stim st im at es w er e ` co nser ns erva va ti ve’ ve ’ in te rm s of D a llin ll in an d Nicolaevsky, but not in terms of the better founded estimates, which were available a t t he he t im im e, e , a nd nd w hi h ic h h a ve ve s ub ub se se qu que nt nt ly ly be en en pr ov ov en en c or or re re ct ct. W e n o ow w h a ve ve summary and detailed archival data which de®nitively prove that the lower estimates were correct.1 5 As explained above, Conquest oscillates in his attitude as to whether to ac cept ce pt th e cu rr ent en t ev al uati ua ti ons on s or to st ay w ith it h hi s old ol d one on e s. Conquest’s estimate of 7 million arrests derived from several sources, including the r ep e p o r ts ts o f A vt vt o rk rk h haa no no v a nd nd D ed e d ij ij er er , a s w el e l l a s t he he a rg rgu m e nt nt t ha ha t t h e p ri ri so so n po pula pu la ti on in 1937 19 37 ± 38 w as la rg er th an the th e 80 0 00 0 that th at w e re ther th er e in M a y 19 34 , an d th at pr ison is on ers er s on av er age ag e stay st ay ed in pr ison is on thr th r e e to f our ou r m onth on thss . 1 6 The T he se ® gu r es are ar e much higher higher than those given by Kruglov and Shvernik Shvernik in their 1954 and 1963 reports to K hr us h che ch e v (s ee be low lo w ) . U nf or tu nate na te ly w e do no t ye t ha v e ac ce ss to the th e de ta iled ile d data upon which these reports were calculated and consequently there remain some grou grounds nds for un unce cert rtai aint nty y conc concer erni ning ng what exact exactlly they they cove cover. r. However However,, as I will will explain below, there is far less basis for accepting the alternative ®gures based on
VI CT IM S O F TH E SO V IE T SE CR E T PO L IC E
319
Conquest’s literary sources, that have already been shown to be very unreliable in estimating the size of the camps population. Conquest’s estimate of the number of executions was also derived from a number of diff differe erent nt sou ource rces. s. He no noted ted that that the the memoir so sour urce cess normall ormally indi indica cate ted d 10% execu executi tion onss to arres arrests ts,, which which coup coupled led with with his his seven seven millio million n arres arrestts would would give give 700 000 execu executi tio ons. Avtork Avtorkhano hanov v est estimated mated 50 500 0 00 000 0 execu executtions ions in thes thesee years. ears. Ginzburg’s account of 70 executions a day in Lefortovo in August 1937 provides the basis for an estimate of 40 000 in Moscow for the two years of 1937 and 1938, which would would give iv e a tot tot a l of 800 80 0 000 000 for for the the coun countr try y as a whol whole. Data Data on mass mass grav graves es unea uneart rthe hed d by the the German Germanss in Katyn Katyn and and Vinn Vinnn nitsa itsa were wer e als also mentio mentione ned d. Other Other sources gave ®gures as high as three million executions, and Conquest concluded: `It will will be seen een that hat no exac exactt esti estimat matee of total otal execu executtions ions can can be made, made, but that that the the 17 nu m be r w as m os t pr ob a bly bl y som so m et hi ng ar ound ou nd a m illi il li on’. on ’. Despite the highly doubtful source of Conquest’s calculations, the ®gure that Conquest proposed has turned out to be m uc h c los lo s er to the th e on e gi ve n by K r uglo ug lo v a nd S hv er n ik th an m an y of Co nque nq ue s t’s t’ s critics, including myself, had expected. Conc Concern ernin ing g mortal mortaliity rate ratess in the the camps, camps, Conq Conques uestt cite cited d Ekart’ Ekart’ss esti estimate mate that that `during the ®rst year about one third of the prisoners die’. In combination with the ®gure of seven million arrests over 1937±38 (less one million executed), this rate of one-third new intake was assumed to provide the two million deaths in the camps in 1937±38. Conquest also cited a work by Wiles to claim an average camp mortality rate of 10% per year, rising to 20% in 1938. He then combined his `high’ population ®gures for the camps with a high average mortality rate of 100/1000 for the period 1936±50, in order to estimate a level of 12 million as labour camp mortality over the 1936±50 period. 1 8 With the deportation of Solzhenitsyn from the Soviet Union and the appearance of hi s Gulag Arkhipelago in the early 1970s the scene was further transformed, with big camp ®gures becoming even more popular. Solzhenitsyn had challenged the Soviet authorities to prove that there were less than 10±12 million prisoners in the labour camps in 1941. Solzhenitsyn’s work inspired a somewhat clumsy attempt by Steven Ros R ose® e® e lde ld e to r ec onst on strr uc t S ovi ov i e t ec ono on o m ic hist hi stor or y by in co r p or at ing in g lar la r ge la bou bo u r cam ca m p 19 ®gures in 1981. Thi T hi s led le d to a leng le ng thy th y de bate ba te w ith it h D avie av iess a nd m e in Slavic Review 20 and with me in this journal, as we tried to correct these ill-informed but politically po pula pu la r vi ew s. Ro se ® el de r e dis di s cove co ve r e d D al lin an d N icol ic olae ae vsky vs ky a nd a dvoc dv oc a t ed a r e tur tu r n to labo la bo u r camp ®gures of over 10 million. I repeatedly insisted that I could see no good reason 21 for moving away from the Jasny, Timasheff, Bergson and Eason ®gures. Conquest commented that, although he had problems with Rose®elde, he preferred his ®gures 22 to m in e, a nd re s tated tat ed his hi s ca se f or ei ght gh t m illi il lion on in the th e cam ca m ps in ea r ly 1939 19 39 . It was at this his time, ime, and and in defen defence ce of his his high high ®gur ®gures es on l abour abour camps camps,, that that Conq Conques uest t launched launched at m e the colourful colourful but totall totally y unfounded unfounded criticis criticisms ms that I have already already cited cited 23 above. The T he se a rg um en ts bec be c am e f ur ther th er co nf used us ed by a som so m ew ha t am a te ur ish 24 discussion of the general level of excess mortality over this period. Thi T hiss is a lar la r ge 25 separate question, which will be dealt with elsewhere. In November 1989 2 6 the th e S ovie ov ie t hist hi stor or ian ia n Z e m sko sk o v pu blis bl ishe he d a set se t of da ta on the th e scale of the labour camps, colonies and special exiles, which included a summary of
32 0
STEPHEN G. WHEATCROFT WHEATCRO FT
annual labour camp population movements for 1934±47. These data showed a total ®gure for the labour camp population of 0.5 million for 1 January 1934, 1.5 million on 1 January 1941 and 0.9 million for 1 January 1953. The population in the prisons and and labo labour ur colo coloni nies es acco accoun unte ted d for a furth further er 500 000 and and there there were were jus ustt und nder er a million in the places of `special’ (forced) exile. These data appear not to have been available to Conquest when he was writing his Reassessment (published in 1990), othe otherwis rwisee it would be dif®cu if®cult to see ho how w he could could have have clai claimed med that that curren currentl tly y available data were supporting his ®gures. 27 I cite cited d thes thesee Zemsko Zemskov v ®gure ®guress in thi thi s jour journa nall in April April 1990. 990. I had origina originall lly y int intended ended to write write a full-l full-len engt gth h arti articl clee on repr repres essi sion on and and the the camps, camps, but I had had jus ust t gained gained acces accesss to the the Soviet Soviet statis statistica ticall archives archives which which contai contained ned TsUNKhU demodemographic graphic materials materials for the the famine famine period, period, and was natura naturall lly y interest interested ed to get those those materials materials into print. print. I kept kept the the same title title,, but compressed compressed the discu discussi ssion on of `New evid eviden ence ce on the scale scale of labo labour ur camps camps and and exi exiles les’ to a page. page. N ever everth thel eles esss, in the the introduction I stated that Th e aca dem de m ic de bate ba te co ncer nc erni ning ng t he sca le of r epr ep r essi on and an d exce ex cess ss m o r tali ta li ty in the th e US S R glasnost’ in the during the 1930s has been raging inconclusively for decades. The spread of glasnost’ USS U SS R has ha s so f ar do n e l ittle it tle t o d am pen pe n the th e at titu ti tu des of the th e r iv al c ont on t ende en ders rs i n this th is d ebate eb ate in the th e W est. es t. B oth ot h Rob Ro b er t Co nqu nq u est and an d m ysel ys elff have ha ve r epea ep eated tedly ly clai cl aim m ed that th at the th e new ne w evid ev iden en ce appeari appearing ng in the Soviet Soviet Union Union has supporte supported d our con¯ictin con¯icting g claims. claims. Conques Conquestt is clearly clearly impres impresse sed d with with the the bulk bulk of litera literarry evide evidenc nce, e, which which does does indee indeed d tend tend to agree agree with with his his conclusions; much in fact is based upon his own work. My attitude has always been to try to ev alu al u ate the th e natu na tu re of t he ev iden id ence ce,, to ch eck its it s o r igin ig inss an d the th e m e thod th od o f ar gum gu m e ntat nt ation ion;; in these terms the evidence that has been appearing in the Soviet press has been very mixed. ¼ In recent months especially there have been tremendous breakthroughs in the availability of archival materials in the Soviet Union.
After After a brief brief discus discussi sion on of Zemskov’s Zemskov’s ®gures ®gures and those of Nekraso Nekrasov, I point pointed ed out th at the th e se ® gu re s ga v e a m ax im um nu m be r of 2.53 2. 53 m illi il li on pr ison is on ers er s in the th e ca m ps, ps , colonies colonies and jails jails,, 2.75 million million special special exiles exiles ( spetsposelentsy) and 65 332 in exile or banishment, which gave a total for 1953 of 5.35 million. `These ®gures are, of course cons consiiderab derably small smaller than han tho hose se cite cited d by Con Conqu quest est and and Rose®e Rose®ellde for the the Gulag Gulag po pula pu la ti on al one’ on e’ . T he cam ca m p m or ta lit y ® gu r es th at c oul ou l d be c alc ulat ul at ed f r om the th e Zemskov data indicated an average level of 70 per thousand for the 1934±47 period. When applied to the smaller level of one to two million in gulag for 15 years, they would would acco account unt for abo about 1.6 millio million n deat deaths inst instea ead d of the the 12 milli million on clai claimed med by Conquest. Great Terro Terror: r: A It was in this same year of 1990 that Conquest produced The Great Reassessment . This must have been a very strange strange year for Conquest Conquest.. On the popular popular level, he was now receiving mass adulation from within the Soviet Union, while at th e ser se r iou io u s ac ad em ic le vel ve l da ta w ere er e ap pe arin ar in g th at cl ea r ly un de r m in ed the th e de ta il of hi s ar gu m en ts. ts . Co nq u e st na tur tu r ally al ly pr ef err er r ed to co nc en trat tr atee on the th e big bi g issu is sues es,, w hic hi c h were gaining him popular adulation. He failed to make any response to his academic crit critic icss in his his book book and and other ther writi writing ngss of the time, time, other ther than than to paro parody dy them hem as 28 `Neo-Stalinis `Neo-Stalinistt Revisio Revisionis nists’. ts’. It is cert certai ainl nly y corre correcct that that the the new mate mate rial rial that that emerg emergee d from from the the Sovi Soviet et Unio Union n at that that time time under undermin mined ed the the trad tradit itio iona nall Sovi Soviet et,,
VI CT IM S O F TH E SO V IE T SE CR E T PO L IC E
321
TA T A BL E 1 C ONQUEST ’ S O RI GI NA L
I n j a il o r c a m p a l r e a dy J a n u a r y 1 9 3 7 A r re st e d Ja n u a r y 19 3 7 ± D ec em b er 1 9 3 8 E x e c u t ed Di ed i n c a m ps 1 9 37 ±3 8 I n c a p t i v it y l a te 1 9 3 8 o f w hi ch i n pr is on i n c a m ps Tot T ot al m ort or t al ity in cam ca m ps 19 36 ±50 ±5 0 A v er a g e l a b o ur c a m p p op u la ti o n A v e r a g e m o r t a l i ty Tot T ot al Ca m p m or tal ity
AN D
R EVISED C ASUALTY F IGURES (M ILLIONS ) 19 6 5
1 9 90
1 99 1
5 7 1 2 9 1 8
? 7 1 2 8 1 2± 4
2 .7 5 7 1 2 ? 1 7 ( o r e ve n a l i t t le l e ss)
8 1 0 0 / 10 0 0 12
7 10 0 /1 0 0 0 10.5 10 .5
5 .5 1 0 0/ 10 00 8
So urce urc e s: 1965: R. Conquest, Conquest, The Great Terror (1968), p. 708. 1990: R. Conquest, Conquest, The Great Terror: A Reassessment (1990), pp. 485±486. 1991: R. Conquest, Conquest, Soviet Studies , 43, 5, 1991, p. 951.
semi-Stalinist view. But it is wrong to suggest that it undermined the view held by th e m or e se r io us of Co nque nq ue s t’s c r itic it icss in the th e W est. es t. V ik tor to r D an ilov il ov,, w ho c our ou r ag e ou sly sl y tr ied ie d to m ak e th is poin po in t in a le tte r to the th e ed itor it orss of Voprosy istorii , was mercilessly 29 attacked for daring to do so. In these heady times Conquest was not prepared to accept the views of his Western critics. And there was much confusion about his clai claim m that hat the the evid evidenc encee beco becoming ming avai availa labl blee in the Soviet Soviet Union Union was tendi ending ng to con®rm his position. However, in one place in his book he did seem to accept that th e ne w da ta w ere er e in dica di ca ting ti ng tha th a t his hi s ear ea r lier li er esti es tim m at es we r e ex ce ssive ssi ve,, w he n he w r ot e th at ins in s te ad of his hi s ea rl ier ® gu re s of f ou r to eigh ei gh t m il lion li on ` the th e tr ue ® gur gu r e m ay be low lo w e r, 30 in the 2±4 million range’. Unf U nf or tu na tely te ly th is w as a r el at ively ive ly r are ar e m om e nt, nt , w hic hi c h was totally outweighed by his general conclusions. In his conclusions, in the same book, he retreated from that position and went only as far as to suggest that he was no w `i nc lin ed to r ed uce uc e th e 8 m illi il li on [ la bou bo u r cam ca m p po pula pu la tion ti on ] at the th e en d of 19 3 8 to 7 milli million on,, or even even a litt littlle less less’. ’. 31 And he maint maintai aine ned d his alleg allegiiance ance to the seven seven million arrests, the one million executions and the two million additional deaths in the camps. Conquest replied to my `More light’ article in late 1991 with a brief article of which about a half was devoted to the excess mortality calculations, and a half to his ®gures on the camps and repression. Here he admitted that his estimates of the scale of population in the camps before and after the Ezhovshchina was likely to have been inco incorr rrec ectt and and that that ins instea tead of ®ve ®ve milli millio on in the the camps camps,, colo coloni nies es,, pri priso sons ns and and spetsposelentsy in January 1937 there may only have been 2.75 million. But he still advocated a ®gure of seven million arrests in 1937±38, and appears still to want to maintain that there was a level of death in camps or by execution of an estimated two to thr th r e e m illi il li on. on . T his hi s w oul ou l d c onse on se qu e ntly nt ly le ave av e 5.5 5. 5 to 6.5 6. 5 al ive iv e r at her he r th an the th e e ight ig ht million earlier estimated (actually nine if prisons are included). However, since his camp ®gure had fallen he would have to increase his estimated level of executions or mortality rates in order to get these results. This would appear to force him to deny th e tr uth ut h of the th e da ta upo up o n w hich hi ch hi s esti es tim m at es init in itia iall lly y de pe nd ed . H e did di d no t ad dr ess es s
32 2
STEPHEN G. WHEATCROFT WHEATCRO FT
ho w he in tend te ndee d to re s olve ol ve this th is ap pa re nt co ntr nt r a dict di ct ion. io n. I nste ns te ad he m ove ov e d on to m ak e th e f ollo ol low w in g am az ing in g c la im : but it remains the case that over the post-1937 period the `literary’ evidence (supported by recent recent Sovie Soviett evide evidenc ncee from from provin provinci ciaal archiv archivees too) too) is consi consider derab ably ly supe superio riorr to the the 32 supposedly rigorous tables cited by Wheatcroft.
Thi T hi s sta tem te m en t s eem ee m s not no t onl on l y ungr un gr ou nde nd e d bu t in di r ect ec t c ontr on traa dict di ct ion io n to w ha t m os t of the more serious evidence was showing. Concerning mortality, Conquest indicated that he was not changing his views when he ar gu ed : It will be seen from the above that however the pre-1937 excess dead are allotted, they can har h ar dly dl y h ave av e to t all ed l ess t han ha n abo ab o ut 1 0 m ill il l ion; io n; and an d th at ( sin ce w e are ar e tol to l d, in ever ev ery y sour so ur ce [sic] that no more than 10% of those in custody in 1937±38 survived), some 8 million more must have then followed. We are already already in the region region of 18 million million without without takin taking g into into account account the victims victims of the post p ost - 193 19 3 8 arr ar r ests est s an d depo de porr tati ta ti o ns ¼ ` T he T went we nty y M ill il l ion’ io n’,, as t hey he y are ar e n ow of ten te n wri wr i tten tt en of in the USSR, cannot be a substantive exaggeration: Wheatcroft rightly remarks that high Soviet ®gures are sometimes derived from my own: but many are not, e.g. those of Shmelev, Yurasov, Mikoyan, Medvedev.
His overall conclusion was confused, and was clearly far more pitched at his appeal to a popu po pula la r au dien di en ce tha th a n to a ser se r iou io u s a ca dem de m ic au dien di en ce , al thou th ou gh he ad dr essed es sed it to the th e latt la tter er.. So far, it is true, I and those who take my view have only been proved right about the facts of the Stalinist terrors, against various Western critics. Still, perhaps even that should be ta ken ke n into in to acc ount ou nt in co nsi der de r ing in g o ur i nevi ne vi tably tab ly le ss e xact xa ct,, bu t not no t ther th er ef ore or e less le ss ser se r iou io u s, consideration of the ®gures, or ranges of ®gures, now under debate.
His ®rst sentence would only make sense if he substituted the words `Soviet Stalinist’ for the word `Western’. His views have certainly not been proved correct against my critici criticism sm of him. Conquest appeared to admit that the 1937 census ®gures gave a m ore reliable indicator of the scale of the camps, colonies and prisons than his estimates based on literary sources. But he failed to notice that those census ®gures were totally compatible with the archival series and would tend to con®rm the reliability of the 33 archival data. The T he kontingent ®gure from the 1939 census is also compatible with th e cam ca m p arc ar c hiva hi va l da ta, ta , bu t Co nq u e st insi in si sts that th at his hi s li ter te r ary ar y s our ou r ces ce s are ar e be tt er. er . A nd pr o ve d ’ (his italics) him despite all this Conquest dared to claim that the data have ` pro correct. In Octob October 19 1993 93 the archi archiv val ®gure ®gures, s, which which clea clearl rly y con contradi tradict cted ed the Conq Conquest uest pi ct ur e, w ere er e pu blis bl is he d in a n im po r tant ta nt ar ti cle by G et ty, ty , R itte it ters rs por po r n & Z em sko sk o v in American Histor Historic ical al Review Review . Natur th e in ¯ ue ntia nt ia l American Natural ally ly thes thesee aut autho hors rs did did not waste waste much time in looking seriously at Conquest’s arguments. Conq Conques uestt repli repliee d to Gett Getty, Ritt Ritteerspo rsporn rn & Zemsko Zemskov v in Ju June ne 1994 and and agai again n appeared in some places to accept that his earlier estimates of the size of the camp po pula pu la ti on ha d be en to o hi gh . H e trie tr ied d to ex pla pl a in th e err er r or in te rm s of the th e le vel ve l of deaths and executio executions ns being higher than than he had earlier earlier assumed. assumed. `Generally `Generally speaking, speaking, over the whole period period,, Western Western ªhighº ªhighº estimat estimates es overes overesti timate mated d camp camp popul populatio ations ns
VI CT IM S O F TH E SO V IE T SE CR E T PO L IC E
323
pa r tly be ca use us e w e unde un de r e stim st im at ed ex ec utio ut ion n s and an d ot he r dea de a ths’ th s’ . 34 A n d h e w as as insistent in again claiming the superiority of his literary and unsubstantiated journalistic sources. Ri tter tt er spor sp or n & G et ty w ere er e quic qu ic k to r ep ly to C onqu on quee st as f ollo ol low w s: he h e h as p r esente ese nte d a f amil am ilii ar m e nu of p r ess art ar t icles ic les with wi th sen sati sat i ona l asser ass er tion ti onss f r om unveri®able sources. We would be glad to see a single exact citation from such sources, whose whose nature nature we apparent apparently ly should should trust trust becaus becausee a small small number of post-So post-Soviet viet of®cial of®cials claim to have seen them. Nothing should prevent the Russian government from putting such dataÐeven if they existÐat the disposal of researchers. 35 Although the unsubstantiated press statements Conquest cites reveal a lot about the imagery to d ay’s ay ’s Russi Ru ssi an cit izens iz ens h ave av e of th eir ei r ow n c oun ou n try’ tr y’ss past pa st,, t hey he y co nst it ute sou so u rces rc es o n the th e hist h ist ory of m enta en tali li ties an d i ndic nd ic ate no thi th i ng a bou bo u t pe nal na l r epr ep r ession ess ion in the th e U SS R bey be y ond on d its it s impact on people’s minds. Times are changing, but the nature of Conquest’s sources and the way that he employs them make him a prisoner of the self-image of the society he seeks to describe. It is astonishing that at the moment when we ®nally have massive internal documentationÐ more detai detaile led d than than anyth anythin ing g the the Nazis Nazis left leftÐsch Ðschola olars rs would would conti continue nue to spec specula ulatte on altern alternaative tive reali realiti ties es and and not occup occupy y thems themselv elves es with with the exis existi ting ng volumin voluminous ous recor records ds.. Speci Special alis ists ts in the the Frenc French h Revol Revolut ution ion wast wastee littl littlee time time argui arguing ng with with write writers rs who limit limit th em selv se lves es to qu otin ot ing g w ha t r esp ect able abl e p olit ol it ician ici anss an d jou jo u r nali na lists sts p rete re te nd to kn o w ab out ou t the th e subject. It is testimony to the sad state of their trade that students of Soviet history are not in a position to follow the example of their colleagues in other ®elds.
I share their astonishment. In fact four years later the grounds for astonishment are even greater than they were in 1992±94. Much more material has now appeared from th e a r ch ives iv es,, a nd m or e W e ster st ern n sc hola ho larr s a r e be c om ing in g f am il iar ia r w ith it h th ese es e m a ter te r ials ia ls.. But curiously many in¯uential books are being produced for the general readership th at ar e co nt inui in uin n g to subs su bscc r ibe ib e to the th e f an ta sy th at ` lite li terr a ry sour so ur ce s’ w hi ch c on® on ® r m Conqu Conquest est’s ’s view view are su supe peri rior or to the mass mass archi archiva vall materia materials ls.. The recen recentt genera generall European History by Norman Davies is a quite extreme example of thisÐso extreme 36 th at I ca nnot nn ot re sist si st putt pu ttin in g th at re co rd s trai tr aigh gh t a s w e ll. ll . Far F ar f r om he lpin lp in g th e dis di s cipl ci plin inee to a na lyse ly se and an d im pr ov e ou r unde un de r stan st an ding di ng of the th e ph en o m en a of the th e ter te r r or , Co nque nq ue st co ntin nt in ues ue s to de n y th at the th e ar c hiva hi vall sour so ur ces ce s c an tell te ll us anything about the scale and dynamics of repression, and he is still insisting that hi s old ol d ® gur gu r e s a r e c or r ec t. In his his repl reply y t o my arti articl clee Conq Conques uestt not not only nl y rep repeats eats his old old clai claims ms abou aboutt the the superiority of his unsubstantiated literary and journalistic sources, but also claims that th e ar ch ival iv al da ta c an be ig no r ed si nce nc e they th ey a r e in co m pa tibl ti blee an d cl ear ea r ly unr un r el ia ble. bl e. Thi T hi s is a ve ry ne ga tive ti ve an d r et ro gres gr esss ive s ugge ug ge s tion ti on . I t is f a r less le ss ex cu s able ab le th an Conquest’s nostalgic attachment to his earlier ®gures. However, since he goes on to attempt to explain what he thinks is wrong, it does give me the opportunity to explain why he is wrong and to demonstrate that the major part of his incompatibilities arise from his own lack of understanding of the problem. Let us have a look at these data and at Conquest’s claims regarding their reliability and comparability.
32 4
STEPHEN G. WHEATCROFT WHEATCRO FT
The nature and origins of the different sets of archival data, their comparability and reliability, and Conquest’s suggestion that the archive data are consciously falsi®ed The source and a brief history of the archival data
The T he S ov ie t r e pr ess es s ive s yste ys tem m w as c om plex pl ex a nd r eq uir ui r ed re c or ds a nd a r ec or d-ke d- ke ep ing in g syst sy steem to op operat eratee. The manage managers rs of the the Gulag, Gulag, labo labour ur col colony and sp speci eciaal exil exilee empires needed a set of accounting data to plan their work. The secret police and ju dici di ciaa l au thor th or it ies ie s nee ne e de d to ke ep r ec or ds. ds . T he ce nt r al pa r ty le ad ersh er ship ip al so r eq uire ui red d pe r iodi io dicc re po r ts f r om the th e se cre cr e t poli po licc e/M inis in istr try y of I nter nt er nal na l A ff ai rs on de ve lopm lo pm e nts nt s in the Gulag, labour colonies and special exile areas, and on the policing and mass repression operations. In their time these of®cial records were kept in the appropriate sec secret archi archive vess of the the NKVD and and the the part party y leaders eadershi hip. p. These These secre secrett acco accoun unti tin ng materials should not be confused with the non-secret propaganda materials that were pu blis bl is hed he d at th e tim ti m e. When the Ministry of Internal Affairs (MVD) lost control of the labour camps (ITL) during the Khrushchev period, the labour camp archives were transferred to the State State A rchives rchives of the the Russian Russian Federat Federation ion (GARF) or T sGAOR as it was known known then. then. What Conquest refers to as `the Zemskov ®gures’ are some of the summary data from th ese es e a r ch ives iv es , w hic hi c h the th e Ru ssia ss ian n hi stor st or ian ia n V . N . Z em sko sk o v w a s a ble bl e to publ pu blis is h f ro m th e la te 1980 19 80ss . 37 The T he se r ep r es ent en t a f ew ® gu re s f ro m am on gs t the th e th ousa ou sa nd s of ® les le s of Gulag accounting data which are now freely available for examination in fond 9401 38 of GARF. Th e fondy contain conjunctural reports on Gulag work, supply reports concern concerning ing Gulag Gulag request requestss to have the government government increa increasse averag averagee perso personal nal food food rati ration onss and and the the siz size of the Gulag Gulag con conting tingent ent on food food ratio rations ns,, health ealth repor reportts, and and general accounting reports. T he da ta th a t Z em s ko v pu bl is he d in 19 9 1 c an be f ou n d in th e se ® les le s a nd a r e 39 rough roughly ly but not tot totally ally compati compatibl blee with with the other other Gulag accou account nting ing data. data. Zemskov’s main summary tables showing the annual transfers of population to and from th e se la bo u r c am ps f or the th e ye ar s 19 34 ± 53 a r e give gi ve n in A pp e ndi nd i x 2. T he se s um m ary ar y Gulag accounts appear to be the sum of the accounts for each separate camp unit. T he co m m a nd e rs of th ese es e s ep ar ate at e cam ca m ps pr ov id e d an nu a l da ta on tran tr an sf ers er s of 40 inmates to and from prisons or other places of detention ( mesta zaklucheniya MZ ) , to a nd fr om oth ot h e r ca m ps. ps . T he y li st ed th os e w ho ran ra n a w ay an d th ose os e w ho w e r e recaptured; those who were liberated and those who died; and there was also a small `other section’. Western historians who consider that all these data were falsi®ed 60 years ago, and th en he ld in se cr et to be pr oduc od uc e d in or de r to dis di s inf in f or m the th e m , ap pe ar to be suff su ff er ing in g from an exaggeration of their own importance. When Gulag of®cials were pleading for more supplies they had no incentive to underestimate the number of prisoners. When Gulag of®cials were planning production they needed to know the real number of prisoners. Their health departments needed to know how many were dying. When MVD leaders were brie®ng Stalin in their top security `Osobye papki’ reports they ha d go o d r e ason as on to av oid oi d the th e c ha r ge of m isle is le adin ad ing g him hi m . W he n tw o dif di f f er ent en t gene generat ratio ions ns of MVD of®ci of®cial alss were were brie®ng rie®ng Khrushc Khrushchev hev on the the iniq iniquit uitie iess of their heir pr e de ce ssor ss ors, s, in the th e ir top to p sec se c ur ity re por po r ts, ts , K r uglo ug lo v in 195 19 5 4 and an d S hve hv e r nik ni k in 19 63 , th ey sim si m ila il a rly rl y had ha d m or e to lo se th an to ga in by f al s ifyin ify in g th e ® gu r es . O f co ur se it
VI CT IM S O F TH E SO V IE T SE CR E T PO L IC E
325
would be rash to presume these data were in any absolute way perfect, but there seem to be no intr in tr insi in sicc gr ound ou nd s f or pr es um ing in g th at th ese es e indi in dicc ator at or s a re gr ea tly tl y f al si® ed . Data on arrest arrests, s, sente sentenc nces es,, appeal appealss and execut execution ionss were general generally ly handled handled sepaseparately rately dependi depending ng upon upon whether whether the the crimes crimes were describ described ed as counter counter-revo -revoluti lutiona onary ry crimes against the state (listed in article 58 of the 1926 Criminal Code of the RSFSR) or as civi civill-cri -cri minal minal crimes crimes (lis (liste ted d in othe otherr arti articl cles es of the the Crimi Crimina nall Code Code). ). The inve invesstig tigatio ation n and and process rocessin ing g of Cou Counternter-Revo Revolu luti tion onary ary (CR or in Russi Russian an K-R) crimes were normally supervised by the security agencies, OGPU until 1934 and then NKVD and later MVD. Although counter-revolutionary crimes could be tried by the civil courts, they were more frequently tried by special non-judicial organs, e.g. the Collegium of OGPU, the troiki of of NKVD, Special Councils of the Military Collegium and military tribunals. This was particularly the case in 1937 and 1938. The data on arrests, sentences and appeals of all criminal cases and those few CR cases handled by the courts are available for the years from 1937 in the NKYust ®les in GARF (F.9492s) (F.9492s) and are accessi accessible ble (see (see Appendi Appendix x 2). T he data data on arrest arrests, s, senten sentence cess and appe appeal alss on most CR crimes, crimes, handl handled by the the secu securit rity y agen agenci cies es,, are locat located ed in the the OGPU archives and are generally not accessible. Some MVD reports based on the OGPU data are available in the special ®le (Osobye papki) reports of the MVD to Stalin, Stalin, to Molotov and Khrushchev Khrushchev that that are also available available in G ARF. These include include the Kruglov 1954 Report and the Shvernik 1963 Report (see Appendix 2). Some researchers have been able to consult local security archives and details of 41 arrests, sentences and actual executions are available in these sources. Again these local materials would normally be divided between the different sentencing authori ti es. es . Below Below I will will discu discusss the compara comparabi bili litty of the diffe differen rentt part partss of these hese dat data and and consider consider Conquest’s Conquest’s critic criticis isms ms concerning concerning their reliabilit reliability y and his proposals proposals concernconcerning alternative sources.
The comparability of the archival data on the size of the labour camp population and the data on transfers to the camps
In his his `Comment’ `Comment’ Conques Conquestt no notes tes that that `we are all all incl inclin ined ed to acc accept ept the Zemskov Zemskov to tals ta ls ( ev en if no t as co m pl et e)’. e) ’. Bu t he goe go e s on to gi v e a tota to ta lly m isle is le adin ad ing g interpretation of what these data indicate: with their 14 million intake to Gulag `camps’ alone, to which must be added 4±5 million going to Gulag `colonies’, to say nothing of the 3.5 million already in, or sent to, `labour settlements’. However taken, these are surely `high’ ®gures.
It is a little unclear what Conquest is suggesting here. On the face of it he appears to be pr op os in g that th at w e r ei nsta ns tate te his hi s ea rl ier hig hi g h ser se r ies of la bou bo u r c am p popu po pula latt ions io ns , which he had earlier appeared to abandon. By making reference to a camp intake in 1934±53 as high as 14 million Conquest is clearly trying to breathe fresh life into his earlier ®gures. A net intake of 14 million would seem to allow plenty of space for 7 milli million arrest arrestss in 19 193 37±38, the the high 10 1 milli million onss of deat death hs in the the camps camps that that Conquest still appears to adhere to, and still allow for the addition of 5 1 million in
32 6
STEPHEN G. WHEATCROFT WHEATCRO FT TA T A BL E 2 T HE S I Z E
OF THE
S OVIET L ABOUR CAM P P OPULATION AND C AMPS , 1934±1953 1934±1953 (M ILLIONS ) T o c am p s
THE
O u t o f ca m p s
(a) (a ) Documented archival sources: 1934±53 P r i so n s (M Z ) 8 O t h er c a m ps ( IT L ) 5 .8 Rec Re c ap ture/ tur e/rr un away aw ay 0.2 O th er 0 .0 6 A ll 1 4 .1
1 .2 4 .6 0.4 0 .1 2 6 .3
L i b er at e d D ie d
6 .4 0 .9
C am p p o p u l a ti o n r e c o r d e d i n 1 9 3 4 C am p p o p u l a ti o n r e c o r d e d i n 1 9 5 3 G r o w t h o f c a m p p o p u l a ti o n A l l l ib ib er er at ate d, d, di ed ed , a nd nd g ro ro wt wt h i n po pu pu la la ti ti on on (b) (b ) Conquest high ®gures: 1997 Tra T ra nsfer nsf erss t o cam ca m ps, ps , 19 34± 34 ± 53 L ab ab o ur c a m p po p u l a ti on a v e r a g e A v e r a g e m o r t a l i ty Conques Conquest’s t’s estima estimate tess o off camp camp mortal mortalit ity y
T RA NS F ER S
TO TH E
N e t t o c am p s
6 .8 1 .2 2 0.2 2 0.06 7 .8
0 .5 0 .9 0 .4 7 .7 .7
14 8 1 0 0 / 1 0 0 0 p e r y e ar 12
So urc es : See Appendix 2. Sourc R. Co nq uest ue st,, The Great Te rror rror (1968), p. 710, `T aking the conservative ®gures of an a verage verage over the period period 1936±50 1936±50 inclusiv inclusivee of eight million million populatio population n of the camps and a 10% death rate per annum, we get a total casualty casualty ® gure of 12 million dead’.
th e cam ca m ps . H is vie vi e w s ne e d su ch a hig hi g h ® gu r e of c am p inta in ta ke; ke ; an ythi yt hi ng unde un de r 10 million would clearly challenge their credibility. Whatever it is that he is aiming at, his argument is invalid and can be shown to rest upon his inadequate understanding of how the labour camp statistics were put together and what they show. The ®gure of 14 million is a gross transfer of prisoners from 42 camp to camp and from prison to camp. Th T h e co r rec re c t ne t ® gu re , ha vin vi n g de du ct ed the th e tra tr a nsf ns f er s be tw e en pr ison is onss a nd c am ps, ps , et c. is ab ou t 7. 8 m illi il lion on , w hi ch is cl ea rly incompat incompatib ible le with with Conqu Conquest’ est’ss rather rather than than with our view view (see (see Table Table 2). Conquest Conquest’s ’s critici criticisms sms of my data, on this score, score, follow simply from his own confusion confusion about w hat th e da ta r efe ef e r to . The comparability of data on sentences and transfers to the camps
Und U nd e r the th e he ad in g ` Sh v e r nik ni k r ep or t poin po in t b) ’, Co nq u e st r ai ses the th e sam sa m e m isco is co ncep nc ep ti ons on s ov e r th e da ta bu t no w appl ap pl ies ie s th em to the th e ye ars ar s 1937 19 37 ±38 ±3 8 , and an d take ta kess hi s ar gu m en t a little way further, by bringing in the data on sentences. He argues that there is an inconsistency between the Shvernik report ®gures of 1 372 392 arrests and 681 692 executions in 1937±38 and the Zemskov ®gure of 1 853 513 entering Gulag camps in th at pe r iod io d . C onqu on qu es t im plie pl iess that th at the th e re is a n err er r or of a bou bo u t 1.1 1. 1 5 m illi il li on. on . T he S hv er n ik r ep or t m at er ials ial s li sted st ed no t on l y the th e ar r es ts of thos th osee ch ar ged ge d w ith it h CR 43 offences but also the sentences of CRs to execution, prison and camp, exile and other.
VI CT IM S O F TH E SO V IE T SE CR E T PO L IC E
327
TA T A BL E 3 A RRE ST S , E XECUTIONS AND G RO W TH
IN
G UL AG L ABOUR
IN 1937±1938 1937±1938
(M ILLIONS )
(a) Documented archival data
P r is on s (M Z ) O t h e r c am p s ( I T L ) Re c aptur ap tur e/ru e/ ru n away aw ay O th er A ll
A l l s e n t e n c es E xe c u t i ons T o cam ca m ps a nd pri pr i son
T o c a mp s
O u t o f c a m ps
1 .4 4 0 .4 1 3 0.0 58 0 .0 0 9 1 .9 2
0 .1 0 0 .4 5 4 0.0 9 0 .0 2 0 .6 6
P o l i t i ca l
C ri m i n al
1 .3 4 5 0 .6 8 1 0.6 35
1 .6 5 4 0 .0 0 5 0.7 30
N et
1 .3 4 2 0.04 2 0.03 2 0.01 1 .2 6 All
3 .0 0 .6 8 6 1.3 65
So urc es : see Appendix 2.
The T he se da ta indi in di cate ca te th at 63 4 82 0 CR s w ere er e s en tenc te nced ed in th ese es e ye a r s to the th e pr ison is on s an d camp ca mpss . Of these hese,, 256 25 6 000 were were list isted as bein being g sent senten enced ced to under under thre threee year yearss impri impriso sonmen nmentt, with with the the remai remaini ning ng 378 378 000 prob probab ably ly bein being g sent sent to t he camps camps.. Zemskov’s gross ®gure for all transfers to camps (including transfers between camps and from prisons and other places of detention to camps) was about 1.9 million as indi indicat cated ed by Conq Conques uest. t. But if we were to excl exclude ude the the 413 000 tran transf sfers ers betwee between n camps and the 100 000 returned to prisons, the transfers into Gulag camps would be reduc reduced ed to 1.4 1.4 milli million on.. We are are cons conseq eque uent ntlly fac faced with with expl explai aini ning ng where where the the additional 600 000 prisoners came from. This is the incompatibility that Conquest is referring to. 44 However, if we look at the data on arrests and sentences under criminal charges, which are available in the Supreme Court archives (GARF, F.9492s; see Appendix 2), it is clear that in these these years about about 700 000 people were sentenced sentenced under under non-politic non-political al charges to loss of freedom. Some 180 000 were sentenced to terms of less than one year and so would probably probably have served served these these terms terms in prisons prisons or labour labour colonies colonies,, leaving about 520 000±600 000 who would be the intake of criminal contingent into th e c am ps in th ese es e ye a rs. rs . A ga in w e see se e no m a jor jo r co ntr nt r a dic di c tion ti on , s im ply pl y an othe ot he r re¯ection of Conquest’s limited understanding of what the data refer to. In a similar similar w ay, under the heading `Shvernik report point point c)’, Conquest Conquest claim claim s that `the `the 9.8 milli millio o n giv given by Zemsk Zems kov as ente enteri ring ng Gula Gulag g camps camps in 193 1939±52 is not compat compatib iblle with with the the 1.1 1.1 millio million n ®gure ®gure of arres arrests ts,, minus minus execu executi tion ons, s, the the Shverni Shvernik k report gives for this period’. Again Conquest’s problems would disappear if he only understood the data a little bett better er.. Ins Instea tead of the Zemsko Zemskov v data ata givi giving ng a gros grosss in¯o in¯ow w into nto the the camps camps of 9 .8 million, they give a net in¯ow of 5.4 million. The Shvernik report would indeed only allow one million `politicals’ to enter the camps, but the data on criminal sentences allow allow 7.6 7.6 million million other other prison prisoners ers to be sentenc sentenced ed to loss loss of freedo freedom, m, and of thes these roughly 5.8 million were sentenced to more than two years. This will easily cover the ne t in ¯ ow of 5. 4 m illi il li on. on .
32 8
STEPHEN G. WHEATCROFT WHEATCRO FT TA T A BL E 4 S UM
OF
A LL T RA NS FE RS R EPORTED BY S OVIET L ABOUR C AMPS , 1939±1952 1939±1952 (M ILLIONS )
(a) Documented archival sources
P ri so n s ( M Z ) O th e r c a m ps (I T L ) A l l fr o m p r is on s a n d ca m p s
P o l i t i c al C ri m in al A ll
T o ca m p s
F ro m c a m p s
N et t o ca m p s
5 .2 6 4 .7 0 9 .9 6
1 .0 3 3 .5 2 4 .5 5
4 .2 2 1 .1 8 5 .4 0
A r r e st s
E xe cu ti on s
To c a m ps a n d M Z
1 .1 1 1 2 .6 2 1 3 .7 3
0 .0 5 0 .0 3 0 .0 8
0 .9 9 7 .6 0 8 .5 9
So urce ur ce s: see Appendix 2.
The comparability and reliability of data on executions
Another argument, which Conquest lists as `Shvernik report point a)’, is somewhat diff differe erent nt.. At this this point point Conq Conques uestt impl implies ies that that I had had not not real realis ised ed that hat the the 4 464 executions of CRs in 1939 and 1940 were too low to allow the inclusion of 25 700 Poles shot in 1940. If we ignore for the moment Conquest’s minor confusion over the ®gures of 25 700 and 21 857, 4 5 it can easily be pointed out that I was aware of this pr ob le m an d qu ite it e cl ear ea r ly r ef er red re d to it: it : Th e clea cl earr est in d icati ic ation on of a m ajor aj or om ission iss ion is the th e ex ecut ec utio ion n of 2 1,85 1,8 5 7 Pol Po l es, Belo Be lorr ussia us sians ns and West Ukrainians in 1940 following the Soviet occupation of parts of Poland as a result of the Molotov-Ribbentropp Molotov-Ribbentropp Pact¼. T he scale of such additional killings killings awaits the opening of the Presidential archives. 46
Elsewhere Elsewhere in my articl articlee I referred referred at some length length to prisoner prisonerss of w ar and the foreign foreign int interned erned po popu pulat latio ion. n. Conq Conques uestt is cert certai ainl nly y correc correctt that hat thes thesee ®gures ®gures need to be included in order to make a proper evaluation of the total costs of Stalinism. But he is incorrect incorrect in suggesti suggesting ng that that there is necessaril necessarily y something something suspici suspicious ous about their their exclusion, and that their exclusion is indicative that many similar mass executions were excluded. I think that the conclusions that we should make from this incident are exactly the opposite to those suggested by Conquest. T he da ta on the th e exe ex e cu tion ti on of th e P ole ol e s, th e ex ec utio ut ion n of de ser se r ter s 4 7 and and the the tre tr e atm en t of othe ot he r f or eign ei gn na tion ti onaa ls, ls , dur du r in g w arar - tim ti m e , w ere er e no t in clud cl ud ed in th e ge ne r al ta bl es that th at the th e M V D pr ep ar e d f or K hr ushc us hc he v . M ost os t r e cons co ns tr uc tion ti onss of stat st at istic ist ical al accounts carried out by Soviet of®cials in the post-World War II period used either th e pr e - 193 19 3 9 boun bo unda da r ies ie s or the th e post po st-- 194 19 4 5 boun bo unda da r ie s. T he Ba lts lt s ar e ge ne rall ra lly y incl inclu uded, ded, becaus ecausee they hey were were inclu included ded in the the po post st-Worl -World d War II bou ound ndaries aries of the the US U S SR , bu t th e Po le s a re ge ne r ally al ly e xc lude lu de d. W e w oul ou l d ne ed to tr eat ea t W es ter n U kr ai ne more carefully. But it is totally understandable that these ®gures exclude POWs and th e a r r es ted P ol es. es . A much more signi®cant problem over the comparability and reliability of data on executions concerns the extent to which executions in the localities exceeded those authorised at the centre, and whether the reported ®gures are based on the central
VI CT IM S O F TH E SO V IE T SE CR E T PO L IC E
329
limit limitss or a su summ mm ation ation of local local ful®lments ful®lments.. Khlevnyuk Khlevnyuk has recent recently ly addres addresse sed d this this question in his article `Les mechanismes de la grande Terreur des annees 1937±1938 48 au Turkmenist Turkmenistan’ an’.. On the the basi basiss of a sp spec ecia iall inv investi estiga gati tio on carr carrie ied d out ou t by the the P r oc oc ur ur at at or or ’s ’s o f ® ce ce u n nd d er e r t he he d ir ir ec ec ti tio n o f t he he P ol ol i t bu bu r o a nd nd N K V D U S SR SR i n Decembe Decemberr 19 193 39, it was esta establ bliish shed ed that that there here had had been been at least east 81 812 2 un unau autthorise horised d killings by the Turkmenistan NKVD in 1937±38. This represented roughly 25% more th an th e 322 32 2 5 th at ha d be en au thor th or ised is ed by the th e c en tr e. K hle hl e vnyu vn yu k po int in t s ou t that th at th ese es e ®gures could well indicate that the central ®gures reported under Khrushchev could 49 be `lower than reality’. However, it should be pointed out that Turkmenistan was pa r ticu ti cu lar la r ly di stan st antt fr om M osco os co w , it al so ha d a pa r ticu ti cu lar la r ly a ggr gg r e ssive ss ive loc lo c al N K V D and that these excesses were identi®ed and condemned by the central authorities in Moscow. As I mentioned in my earlier article, Khlevnyuk, Zemskov and Roginsky, `when independently consulted on this issue [the possible under-reporting of execu ti ons on s ] all al l ag r eed ee d that th at the th e of ® ci al ® gu r es f or e xec xe c utio ut ion n s m ig ht re quir qu ir e som so m e c or re c ti on from 800 000 to possibly 1 million. None of them were prepared to accept the 1.5 million ®gure currently being advocated by Conquest’. 5 0
Conquest’s claims regarding the more reliable literary sources
Before considering Conquest’s speci®c claims regarding his literary sources, I would ju st like li ke to m ak e on e poin po in t , a bou bo u t th e gene ge ne r al r elia el ia bili bi lity ty of ey e-w e- w itne it ne ss ev al uati ua tion ons, s, th at em e rg ed f r om m y stud st ud y of the th e H oloc ol oc au s t li ter te r at ure. ur e. It is clear that the quality of eye-witness reports for the Holocaust was immensely super su perio iorr to tho those that that we have have for Stal Stalin’s in’s repre represssion. sion. In many case cases they they were were recor recorde ded d by a prof profes essi sion onal al lega legall service rvice,, whic which h was su supp pport orted ed by t he milit militaary authori authority ty of occupy occupying ing forces forces, which which were active actively ly pursuin pursuing g thes these invest investiga igati tions ons.. The T he se in ve stig st igaa tion ti onss w ere er e ca r ri ed ou t ve r y so on af ter te r the th e ev en ts. ts . T he inve in ve stig st igat at ions io ns he a rd ev ide id e nc e no t onl on l y f r om cam ca m p in m at es bu t the th e te stim st imon on y of cam ca m p c om m an de r s and individuals who held key positions in the terror administration. Nevertheless, it is now accepted that the evidence accepted at Nuremberg to prove that 3.5 to 4.5 million people perished in Auschwitz alone is incorrect and that the true ®gure for 51 Auschwitz mortalities is about 1.1 to 1.5 million. In his comment Conquest chides me for not taking seriously the unsubstantiated ®gur ®g ures es that that he had cite cited d fro from the comment commentss of Mikoy Mikoyan’ an’ss so son, n, Khrus Khrushc hche hev’ v’ss son-in-law, son-in-law, O lga S hatuno hatunovskaya, vskaya, Dm itrii itrii Volkogonov, Volkogonov, Colonel Colonel Grashoven, Grashoven, Aleksandr Aleksandr Yakovlev and other unspeci®ed researchers in the Ministry of Justice archives, who, he tell te llss us, us , `a ll supp su ppor or t th e ªhig ªh igh h º esti es tim m at es’ es ’ an d ` gi ve a r r est es t ® gu r es in the th e 19 ± 21. 21 . 5 million range and death ®gures [presumably Conquest means `execution’ ®gures] of 7 million’. T he se are ar e the th e s am e ca s es th at Co nque nq ue s t had ha d e ar lie r cite ci ted d in his hi s r es po nse ns e to m y article of 1990 and the 1993 article of Getty, Rittersporn & Zemskov in A H R , June 1994. It m ade m ore sense sense to cite cite them then then than it does now. A t that tim tim e the response response 52 of Getty & Rittersporn was appropriate, and it is even stronger now. In Table Table 5 I have ave list listeed the the d ocumen ocumente ted d arch archiv ival al so sour urce cess, toge togeth ther er with with the the undocumented claims, in a manner that should help us see exactly what it is that Conqu Conquest est wants wants us to beli believ eve. e. Some of our coll colleeagues agues lik like Edwin Edwin Bacon Bacon seem seem to
33 0
STEPHEN G. WHEATCROFT WHEATCRO FT TA T A BL E 5
A R RE ST S , S ENTENCES AND E XECUTIONS A CCORDING TO THE A RC HI VE S AN D A CCORDING A PPARENTLY F AVOURED BY CONQUEST ( M ILLIONS ) ( a ) A rre st s a n d se n t e n c es A ll
(i) Documented archival sources 1 9 21 ±5 3 1 9 34 ±5 3 1 9 37 ±3 8 1 9 39 ±5 2 (ii) Undocumented claims Volkogonov 1 9 29 ±5 3 Shatunovskaya 1 9 35 ±4 1 Grashoven 1 9 35 ±4 5 ( b ) E x e c u t i on s
(i) Documented archival sources 1 9 21 ±5 3 1 9 34 ±5 3 1 9 37 ±3 8 1 9 39 ±5 2 (ii) Undocumented claims Volkogonov 1 9 29 ±5 3 1 9 37 Shatunovskaya 1 9 35 ±4 1 Grashoven 1 9 35 ±4 5 General A. Karbainov 1 9 37 ±3 8
A rre st s P o l i t i c al Cri m in a l
A ll
3. 0 1 3. 7
TO THE V ERSIONS S TI LL
S e n t e nc es P o l i t i c al
4. 1 3. 1 1. 34 1. 1
Cri m in al
1 .6 5 1 2 .6 2
2 1 .5 1 9 .8 4 1 8 .0 A ll
0. 68 7 0. 08 3
P o l i t i c al
0. 79 9 0. 74 0 0. 68 2 0. 05 4
Cri m in al
0. 00 5 0. 02 9
7. 2 1. 75 7 7 3. 5
So urce urc e s: Volkogonov, 1929±53: D. Volkogonov, Kuranty , 9 May 1991. Shatunovskaya, 1935±41: Olga Shatunovskaya, Argumenty i fakty fakty , 1990, 22, pp. 6±7. Grashoven, 1935±45: V. Tolz, `Ministry of Security Of®cial Gives New Figures for Stalin’s Victims’, Victims’, RFE/RL Research Report , 1 May 1992, pp. 8±10. Volkogonov, Volkogonov, 1937: D. Volkogonov, Volkogonov, Trotskii , vol. 2, p. 323. Karbainov, 1937±38: D. Volkogonov, Trotskii , vol. vol. 2, p. 323. 53
th ink in k that th at it is po ssib ss ibll e to r ec onc on c ile il e s om e of th ese es e se r ie s. I am doubtful about the value of attempting to do this. T he re seem se em to be no gr ound ou nd s a t all f or taki ta kin n g thes th esee cl aim ai m s s er ious io usly ly.. I t is w ell el l known kn own that that many many indi indivi vidu duals als in the the Soviet Soviet Union Union leapt leapt on Conqu Conquest est’s ’s or Avtor Avtorkk ha no v ’s ear ea r lier li er w ri ting ti ng s , an d w er e co nv in ced ce d tha th a t they th ey w er e tr ue. ue . M a ny of th ese es e indiv individu iduals als appear appear to have been been confuse confused d in thei their discu discuss ssion ionss with with journ journali alist stss, and appe appear ar to have have cite cited d Conq Conque uest st’s ’s or Avtor Avtorkha khano nov’ v’ss ®gure ®g uress as thou though gh they they had alternative sources for them. Conquest appears to have taken heart at the frequent mention of the seven million ®gure, and was quick to claim that this was con®rmation of his his earl ea rliier ®gur ®g ures es.. But other therss are are bound bound to be more more scep scepti tica call . None No ne of thes thesee
VI CT IM S O F TH E SO V IE T SE CR E T PO L IC E
331
individuals has been able to provide any con®rmation of any independent source for 54 th ese es e ® gur gu r es , a nd they th ey ha v e ge ne r al ly no t r ep ea ted te d th ese es e cl a im s.
Additional data on sentences and executions in the military purges
Several Several addit additio ional nal so sources urces of data data have have become become availa availabl blee to indic indicate ate that previo previous us esti estimates of the scale scale of arrest arrests and execut execution ionss in the the Ezhovschi Ezhovschina na were overest overestiimated. Perhaps the most important of these are the materials on the mass purging of th e R e d A r m y a f ter te r the th e T ukha uk ha ch ev sky sk y tria tr ials ls.. During the Khrushchev period it had been claimed that 36 761 military and naval pe r sonn so nn e l ha d be en r ep r esse es sed d be tw een ee n M ay 193 19 3 7 an d S ep tem te m be r 1938 19 38 . 5 5 Conquest accepted such a ®gure in the ®rst edition of his book and claimed that this led to the 56 loss of about 50% of the of®cer corps of 70 000. The T he im plic pl icaa tion ti on w as that th at m ost os t of th ose os e w ho we re r epr ep r esse es sed d pe r ishe is hed, d, al thou th ou gh a ve ry sm all nu m ber be r w er e re ha bili bi lita ta ted te d at the the begi begin nning of the w ar. As late late as 19 1989 89 the dist distiingui nguished shed Russ Russia ian n milit military hi stor st or ian ia n G . A . K um an ev cla im ed in Pravda : According to archival data, for the period of only 27 February 1937 to 12 November 1938 the th e NK V D r ece iv ed f rom ro m Stal St alin, in, M o lo tov to v a nd Ka gano ga nov v ich ic h sanc sa nc tion ti on to sho sh o ot 3 8 679 67 9 m i lita li tary ry per p er son nel. ne l. I f to th is nu m ber be r you yo u add ad d m ore or e th an th r ee t ho usan us and d destr de str o yed na val va l com co m m ande an derr s, and take into account that the destruction of military cadres began before 27 February 1937 and proceeded beyond 12 November 1938, then the number of innocent military commanders who perished would come close to 50 000 and the total number repressed in the army and the ¯eet would be considerably higher.57
However, it is now generally accepted that all these ®gures are very misleading and th at K um an ev ’s ® gu r es are ar e just ju st to tall ta lly y w r ong on g . A r ep or t f ro m de pu ty P eo ple pl e ’s Commissa Commissarr of Defenc Defencee E. A. Shch Shchaden adenko ko did indee indeed d ind ndic icat atee that that 36 761 of®cer of®cerss 58 were discharged from the army in 1937±38, but of these only 10 868 were arrested. Most of the 14 684 of®cers discharged at this time were expelled from the Communist Party for `association’ with those of®cers arrested, and of these roughly a half or 7202 were reinstated by 1939. Of those arrested by the NKVD 1431 were also reinstated by 1939. Overall the total permanent reduction to the military by the purges in these years was more like 17 000, or 22 705 if we include 1939 and other categories of discharge. Since the commanding staff of the army numbered 144 300, rather than th e 70 00 0 ear ea r lier li er be liev li eved ed , the th e pu r ge s a ffe ff e cte d at m os t a bou bo u t 16 % of of ® c er s, an d not no t th e 50 % ea r li er be liev li eved ed . T his hi s is an in tere te re stin st ing g tale, ta le, w hi ch inc in c ide id e ntly nt ly w ar ns ag ains ai ns t th e da nger ng er s of ac ce ptin pt ing g jo ur na list li stic ic r ep or ts f r om r espe es pecc tabl ta blee indi in divi vi dua du a ls w ho ough ou gh t to be in the th e pos po s itio it ion n of knowing what they are talking about.
Finally: a word about the `truth’ and what to expect from data
I was very surprised that Conquest should state that I was claiming to present `the tr ue, ue , ªar ªa r ch iva iv a lº to tals ta ls f or the th e vic vi c tim ti m s of S tali ta li nism ni sm ’. I m ay no t ha v e gr eat ea t s ym pa thy th y with post-modernism, but I would be reluctant to claim to have found the `truth’. Concerning the data on arrests, sentences and executions, I wrote the following:
33 2
STEPHEN G. WHEATCROFT WHEATCRO FT
Althou Although gh the the main main archiv archivees of the the stat statee secur securit ity y organi organissation ationss are are stil stilll unava unavail ilab able le for general scholarly investigation, a few researchers have been given access and have published some some genera generall ®gures ®gures from from thes these archi archives ves.. These These relea release sed d ®gures ®gures provid providee the the follow following ing indications ¼
Later I pointed pointed out some of the om issions issions and inadequacies inadequacies of these data. In particular particular I stated: Apart from these victims of repression listed above there were other categories of victims, which included those whose trials and sentences were not initiated by the secret police and other groups, including the very large group of forced migrants and exiles, which were not included in the above lists ¼ We should also note that the large number of exiled kulaks and deported nationalities are not included in the above list.
When discussing the data on the labour camps, I was far less circumspect, because we do have have dire direcct acce accesss t o the the arch archiv ival al s ourc ourcee s o n the labour abour camp ca mpss and and our knowledge is consequently much more ®rmly based. On p. 1330 I had written about th e ri val va l esti es tim m at es of the th e sc ale al e of th e labo la bou u r c am ps tha th a t ha d be en m ad e be f or e the th e archi archive vess were were op open ened. ed. And I had had con contrast trasteed the the higher igher esti estimat mates es of `Dall `Dallin and and Nicolaevs Nicolaevsky, ky, Schwartz Schwartz and Avtorkhan Avtorkhanov ov su support pported ed by Conqu Conquest, est, Solzhen Solzhenit itsy syn n and Ros R ose® e® e lde’ ld e’ w ith it h th e lo w er e stim st im at es of T im a shef sh efff , Jasn Ja sny y and an d Be r gso gs o n a nd E as on supported by Wheatcroft’. It was in that respect that I had concluded that: Some specialists on Soviet history are ®nding it dif®cult to adapt to the new circumstances when the archives are open and when there are plenty of irrefutable data; they prefer to hang on to their old Sovietological methods with round-about calculations based on odd statements from emigres emigres and other other informant informantss who are supposed supposed to have superio superiorr knowledge. knowledge.
Perhaps I went too far in suggesting that these data presented `irrefutable’ proof of th e sm alle al lerr se r ies of da ta. ta . I un de r e stim st im a ted Co nqu nq u e st, st , w ho ap pa r en tly s till ti ll ho ld s to hi s earlier high estimates of the scale of labour camps, and to even higher estimates of th e sc ale al e of m ass ki llin ll in gs th an he ea rl ier adhe ad he r e d to. to . Conquest argues that we should only consider perfect data and should reject all data th at ar e no t ` per pe r f ec t’. t’ . T his hi s s eem ee m s to be th e basi ba siss of hi s r ej ec tion tio n of th e ar ch iva iv a l da ta. ta . Conqu Conquest est is wrong wrong in su sugg ggest estin ing g that that I cons consiider the the arch archiv ival al data data to be `perfe `perfect ct’’ re¯ections of the `truth’. This is clearly nonsense. What we have to do is to try to assess the limits of the unreliability of all of these data. When I spoke about `plenty of irrefutable data’ I did not mean to suggest that there were unique pieces of data which gave the perfect truthful picture. Rather I had in mind the accumulation of a mass of different types of data (of differing degrees of reliability), which, when taken to ge the th e r ca m e do w n he a vily vi ly in f a vou vo u r of the th e lo w er esti es tim m at es f or the th e s ize iz e of th e labo la bo u r camps. That is now something that should be considered as academically proven, and Conquest was clearly wrong. The situation regarding arrests and executions is slightly less clear. But Conquest’s proposals concerning them are highly unlikely. Let us look brie¯y at some of the other misconceptions that Conquest has raised concerning the famine.
VI CT IM S O F TH E SO V IE T SE CR E T PO L IC E
333
Misconceptions about `control’ over the famine
Conqu Conquest est ini initial tiallly cite cited d evid evidenc ence, e, base based d on the the preli prelimin minar ary y repor reports ts of a high ighly respected Soviet scholar (Danilov), to claim that Stalin and the Politburo accumulated grain stocks during the famine year of 1932/33. According to Conquest the government had accumulated and held 4.53 mln tons of grain reserves at the height of the 1932/ 19 32/33 33 famine famine,, on the eve of the 19 1933 33 harv harves est. t. A deta detail iled ed inve invest stig igat atiion of the the archives indicates that this was a mistaken claim, based on confusing a plan ®gure for 59 reality. T he P olit ol it bur bu r o c er ta inly in ly inte in te nde nd e d to a cc um ulat ul atee gr a in stoc st oc ks in 19 32 /33, /3 3, an d th ey ha d se ve ral ra l goo go o d r e ason as on s to do so: so : (a) the level of stocks at the beginning of the 1932/33 agricultural year was considered to hav h avee b een d ange an gerr ousl ou sly y low lo w and an d lik li k ely t o cau se a b r eakd ea kdow ow n in reg re g ular ul ar supp su ppll y; (b) the government had been trying to accumulate a series of reserves that it felt it would need n eed i n the th e case ca se of a thr th r eat o f w ar . Th ese wer we r e the th e soso - cal led ` m ob il isa tion f u nd’ nd ’ ; a nd (c) the government had been further disturbed by the Japanese invasion and occupation of Northern Manchuria, which was threatening the Russian Far East. The Soviet government were consequently urgently seeking to build up reserve stocks in the major de®cit region of the th e Russi Ru ssi an Far Fa r E ast .
Despite these factors pressing the government to accumulate stocks, and the clearly express expressed ed desir desiree to do so, the the Polit Politburo buro was repeat repeatedl edly y forced forced to issue issue emergency emergency grain supplies (both seed and food) to agriculture and the famished population. This ha s be e n do cu m en ted te d in so m e de tail ta il by D a vies vi es & W he a tc rof ro f t in a f or thco th co m in g 60 article. Becau Because se of the the sensi ensitivi tivity ty of thi this probl problem em the the go gove vernmen rnmentt un unde derto rtook ok severa severall pr oc e dur du r es of dis di s inf in f or m at ion. io n. R ad ek le ak ed to W este es te rn dipl di plo o m a ts an d the th e W es te rn pr e ss that th at pa r t of the th e r ea s on f or the th e gr a in s hor ho r tage ta gess in 1932 19 32 /33 /3 3 w as the th e ne e d to 61 accumulate grain reserves in the East. The T he S ovie ov ie t m ilit il itar ary y arc ar c hive hi ve s r ev eal ea l th at no such stocks were accumulated at the time, hence the need for an agent of disinforma ti on. on . A lso ls o it ap pe ars ar s that th at al thou th ou gh th e P ol itb it b ur o w a s r epe ep e a tedl te dly y gr an ting ti ng f ood oo d an d seed seed aid aid to starvi starving ng regio regions ns througho throughout ut 1932/3 1932/33, 3, it forbade forbade any any mention mention of this. this. It pr e sum su m a bly bl y di d this th is f or tw o rea re a sons so ns:: ( a ) to co ntai nt ain n w ha t othe ot he r w ise is e w ou ld ha v e be en a rush of applications and (b) in order not to let potential enemies know that reserve stocks were not available. As a result of the Politburo not building up stocks, and repeatedly issuing food and seed aid throughout 1932/33, by the end of the 1932/33 agricultural year grain stocks were almost as dangerously low as they had been at the beginning of the year. There were virtu virtuall ally y no reserves reserves and the operati operating ng stock stockss, i.e. i.e. those those current currently ly circul circulati ating ng stocks within the system, were as low as 1.9 million tons, although, as a result of errors, it seems likely that the leadership at the time thought that there were only 1.4 million tons. At the time, this was considered a dangerously low level. Although the collection of the harvest would begin in early July, it would not get into full swing until late July. Even then it would take several more weeks before the new grain was tra tr a nspo ns po r ted a ro un d the th e co untr un tr y, an d ta ken ke n to the th e m il ls f or m il ling li ng.. S o it w oul ou l d pr ob ab ly be in A ugus ug us t be fo r e the th e ne w ¯ ou r co uld ul d be ta ken ke n to the th e ba ke rie ri e s f or ba kin ki n g into bread. This was a critical transition period of maybe a month or a month and a ha lf. lf . T he co untr un tr y w ou ld nor no r m al ly ne ed a m in im um of tw o m illi il li on ton to n s of tran tr an sitio si tion n
33 4
STEPHEN G. WHEATCROFT WHEATCRO FT
grain to ensure that there was no breakdown in supplies and mass starvation, before th e ne w gr ain ai n w as ha r ve ste st e d, pr oc e ssed ss ed an d tra tr a nspo ns po r ted te d into in to th e ha nd s of the th e consumers. This has all been explained in great detail in our article. Conquest, having made the initial error of publicising Danilov’s preliminary ®gure o f a n a cc cc um um ul ul at at io io n o f 4. 5 m il i l li lio n t on on s o f g r ai a in r es es er er ve ve s, s, n o ow w m is i s qu quo te te s o u urr correction, and demonstrates his lack of understanding of the detailed operations of th e syst sy stem em . I ns tea d of ac ce ptin pt ing g th at the th e r e w e r e no t 4. 5 m il lion li on to n s of r ese r ve stoc st oc ks but 1.9 million tons of transition stocks, he misquotes us as having said three million to ns, ns , an d he c onti on tinu nues es to ins in s ist on de scr sc r ibin ib ing g th ese es e as r es er ves, ve s, a s oppo op pose se d to the th e tra tr a nsit ns itio io n stoc st oc ks. ks . H e is w ro n g in s ugge ug ge s ting ti ng that th at th e sys sy s tem c ould ou ld ea sily si ly ha v e be en drained of all of these circulating stocks without wrecking of the supply system and causing massive urban famine.
Conclusion
The T he c oncl on cl usio us io n that th at ha s to be m ad e f r o m a c ar ef ul an al ysis ys is of Co nque nq ue st’s st ’s cl aim ai m s is th at his hi s c r itic it ic ism s are ar e gr ound ou ndll e ss, ss , an d tha th a t he doe do e s no t und un d e rsta rs ta nd th e m ate r ial ia l th at he seek se ekss to r eje ej e c t. While we lack access to the State Security archives and the Presidential archives, th ere er e is co nsid ns idee r ab le m at er ial av ai labl la blee f r om th e ar ch ive iv e s that th at de m onst on strr ates at es that th at the th e scal scalee of the the camps camps was much lower lower than than Conq Conque uest st had had earl earliier clai claimed med.. Thes Thesee materials appear to be comparable with the published reports from the archives that are still restricted, and they strongly suggest that the number of arrests and executions and deaths in the cam ps w ere signi®cantl signi®cantly y low er than those those claimed claimed by Conquest. Conquest. T he alternative sources offered by Conquest are unsubstantiated journalistic reports from individuals who might have been in a position to see the kind of documents that they are referring to. But the vagueness of their reports, the suspicious focus on the seven million ®gure, and the total lack of any con®rmation must lead us to be very sceptical of the value of these sources. My 1996 1996 arti articl clee atte attempt mpted ed to plac placee so some me of the the avai availab lable le data data in a comp compara arati tive ve framewo framework, rk, and argue argued d that that it was the the scal scalee and and natu nature re of the the Sovi Soviet et massmass-ki kill llin ing g operations of 1937±38 which were particularly signi®cant, in terms of a comparison with mass r epression epression in Nazi Germany. T he Gulag is to some extent extent an inappropri inappropriat atee symbol. T he re s houl ho ul d be no doub do ub t th at th e f utur ut ur e of th e ac ad em ic stud st ud y of S ovie ov ie t s ocie oc iety ty during the Stalin period lies with a critical understanding of the wealth of archival sources, and not with the further analysis of unsubstantiated journalistic reports and literary literary sources. sources. 62 The T hese se lite li tera ra ry m ate ri a ls w ill il l a lw ay s be us ef ul, ul , as se c ond on d ary ar y materials, but their time as the sole and dominant source is over. And we should not be hinder indered ed by the the comments comments of the the maste masters of the lite literar rary y so sourc urces es,, who do not understand understand these these other materials materials.. W e are still still at early stages stages in understanding understanding the nature of Stalinism, Stalinism, but assessing assessing the scale of several of its dimensions and placing that scale into some kind of perspective seem to me to be a useful way to start. I welcome constructive comments to help us improve our understanding, but I reject attempts to limit and cripple our discipline. University of Melbourne
VI CT IM S O F TH E SO V IE T SE CR E T PO L IC E
335
Appendix 1: Corrections to the Kolyma data, and further discussion on the comparability of camp data
In my 1996 article the data which I cited for prisoners arriving at Nagoevo, the only tra tr a nsit ns it po in t into in to K oly ol y m a, c am e fr om the th e So vie vi e t ar c hive hi vess in dir di r e ctly ct ly f r om A. K oz lov lo v , th e cu r at or of th e M ag ad an H isto is to r ical ic al M us eum eu m , as ci ted te d by J . J. S teph te ph an in hi s boo bo o k The Russian Far East: A History . It now appears that either Kozlov or Stephan had mis-cited these data, which refer not to prisoners arriving at Nagoevo but to the total stock of prisoners in Kolyma. T ab le A 1.1 1. 1 co m pa r es th e da ta e ar lier li er ci ted te d f r om P ilya il ya sov so v ’s stud st udy y of the th e size si ze of the th e Dalst Dalstro roii labour abour forc forcee with with thos thosee give given n by Bats Batsaaev from from local ocal arch archiv ives es,, and and by Zemskov from from centra centrall archiv archives. es. T he Batsae Batsaev v data, data, cited cited here here from Tkacheva Tkacheva,, also also pr ov id e in dica di ca tor to r s of m ove ov e m en ts of pr ison is onee r s bo th to N a goev go ev o an d al so th eir ei r r et urn ur n from Nagoevo (the main port of Kolyma in Dalstroi). T he m ain ai n di scr sc r ep an cy be tw een ee n the th e co nt en ts of the th e sepa se parr ate at e Gu la g ® les le s an d the th e pu blis bl is hed he d tota to ta ls give gi ve n by Z em sko sk o v appe ap pe a r s to be the th e om issi is sion on in m os t of the th e G ula ul a g ®les of information from USVITL, which was technically a part of Dalstroi. As I understand understand the situati situation on Dalstroi Dalstroi and USV ITL had a unique relation relationshi ship. p. D alstroi alstroi w as a state enterprise that was taken over by the OGPU and USVITL was a labour camp th at w as tra tr a nsf ns f er r ed to it. it . T he U SV I T L m ate r ia ls (ac (a c co unti un ting ng data da ta,, co njun nj un c ture tu re reports and possibly health materials) were presumably kept in the Dalstroi ®les rather th an in ce ntr al G ulag ul ag ad m inis in is trat tr atio ion n . F or tu n ate at e ly loca lo ca l D alst al st r oi m at er ial s a r e av ai lable, and from them it would appear that much of the difference between the Gulag to tals ta ls by ye ar a nd by c am p that th at I ha v e se en in the th e a r ch ives iv es,, an d the th e Z e m s kov ko v to tals ta ls,, is attributable to the inclusion of USVITL and possibly a few other minor camps. The difference in any case is not great, as can be seen from Table A1.2. It should also be po inte in te d ou t that th at bot bo t h thes th esee ser ies are ar e co m pa tibl ti bl e w ith it h the th e s pec pe c ial co ntin nt inge ge n t r e por po r ts in the 1937 and 1939 censuses. My conclusions are that the Dugin data for 1931±35 may be a little low in comparison with the archival data; the data for 1936±38 are pr ob ab ly abou ab ou t r ight ig ht,, bu t e it her he r th e D ug i n da ta f or 1939 19 39 ±40 ±4 0 are ar e too to o hi g h or the th e Dalstroi and other data are too low.
33 6
STEPHEN G. WHEATCROFT WHEATCRO FT TA T A BL E A 1.1 1. 1 D IFFERENT S ERIES
OF
A RC HI VA L D A T A
A l l La b ou r
1 9 32 1 9 33 1 9 34 1 9 35 1 9 36 1 9 37 1 9 38 1 9 39 1 9 40 1 9 41
OF THE
D ALSTROI L ABOUR F ORCE ,
P ri so ne rs
P il il ya ya so so v
T ka ka ch ch ev eva
14 27 37 48 67 74 144 197 237 246
13 30 35 50 73 92 11 3 18 9 21 6 21 0
000 000 700 700 500 700 600 300 300 900
MOVEMENT 1932±1941
ON THE
05 3 78 2 99 5 30 1 15 0 25 8 43 0 82 6 42 2 67 4
P il il ya ya so so v
31 42 56 63 1 22 1 32 1 42 1 49
8 00 8 00 9 00 0 00 0 00 0 00 4 00 6 00
B at at sa sa ev ev
99 0 0 27 400 32 300 44 600 62 700 80 300 93 900 163 500 176 600 148 300
Z e ms msk ov ov
36 48 70 90 138
31 3 74 0 41 4 74 1 17 0
A rr rri va va ls ls
D ep ep ar artu re re s
Ne t a rr rri va va ls ls
1 3 87 23 7 03 15 6 73 23 2 68 41 3 11 41 6 63 68 2 69 70 4 92 47 3 79 22 9 63
8 72 5 9 74 9 0 12 9 5 50 18 5 23 21 2 48 34 4 92 26 1 76 3 8 72 14 0 66
515 17 7 2 9 66 6 1 13 7 1 8 22 7 8 8 20 4 1 5 33 7 7 7 44 3 1 6 43 5 0 7 88 9 7
So urce urc e s: A. N. Pilyas Pilyasov, ov, Dinamika promys promyshlennog hlennogo o proizvods proizvodstva tva v Magadanskoi Magadanskoi oblasti oblasti (1932±1992gg), (1932±1992gg), Part 1 (Magadan 1993), p. 225. I. D. Batsaev, `Kolym skaya gryada arkhipelaga Gulag, zaklyuchennye’, in Istoricheskie aspekty Severo-Vostoka Rossii: Ekonomika, obrazovanie, Kolymskii GULag (Magadan, 1996), p. 5 0. B. Ze mskov, `Zaklyuchennye v 30-e gody (dem ogra®cheskii aspekt)’, aspekt)’, Sotsiologicheskie issledovaniya , 1996, 7, p. 6, citing GARF, F. 9414, op. 1, d. 1155, l. 2, 20-2.
TA T A BL E A 1.2 1. 2 F ITTING D ALSTROI D ATA
A l l G u l ag a rch iv es (c o oll u m mn n 1)
1 Ja n u a r y 1 9 30 1 Ju n e 1 9 3 0 19 3 1 19 3 2 19 3 3 19 3 4 19 3 5 19 3 6 19 3 7 19 3 8 19 3 9 19 4 0 19 4 1
12 8 15 6 26 0 26 7 43 6 58 3 73 8 78 5 78 6 87 6 98 8 1 10 5 1 29 4
96 3 00 0 00 0 00 0 00 0 00 0 00 0 00 0 10 0 95 7 33 3 23 8 76 5
INTO THE A RC HI VA L
S UM MA RI ES
Dugin less less a r c h i v es D a l st r o i O t h er D u gi n ( c o lu m n 3 P il y a so v ( co lu m n 5 , ( co co lu lu m mn n 2 ) c o l. l. 1 2 col. col. 2) (colu (column mn 4) col. col. 3 6 col. 4)
1 79 0 00 2 12 2 68 3 34 5 10 7 25 8 39 8 20 9 96 1 3 17 1 3 44 1 5 00
0 00 7 00 3 00 3 07 4 83 4 06 8 81 3 67 1 95 4 08 5 24
48 000 17 0 0 2 101 700 2 72 693 2 12 517 54 4 06 34 7 81 1 19 4 10 3 28 8 62 2 39 1 70 2 05 7 59 2
31 800 42 800 56 900 63 000 122 000 132 000 142 400 149 600
4 0 00 0 2 30 000 2 2500 2 9 00 0 3 00 0 19 7 00 0 19 7 00 0 5 6 00 0
So urc es : T he All Al l Gula Gu la g ar ch ival iv al ser ies ie s in co lum lu m n I r ef ers to t he de t ai led m at er ials ial s in GA RF , F. 94 14 , op. 1, d. 2919. T he D ug in ser ies ie s wa s ®r ® r st pu bl ished ish ed by A. Dugi Du gi n , `G ul ag gl az am i ist or ik a’, a’ , in Na Na boevom postu and was reprinte reprinted d in Soyuz , 9 February 1990, p. 16 and referred to ®gures from a Kruglov, Ru de nk o & Go rshen rsh en in re po rt to Kh ru shc hev he v in 1954 19 54 . T he l oc al Da lstr lst r oi serie ser iess is t aken ak en fr om Pil Pi l yasov ya sov , see Ap pe ndix nd ix 1.1 .
VI CT IM S O F TH E SO V IE T SE CR E T PO L IC E
s e l b a t e e S . 6 5 9 1 ± 0 2 9 1 , R S S U e h t n i s n o s i r p n e e w t e b d 3 . n 2 a A m d o r n f a , , o 2 . t 2 s r e A , f 1 s . n 2 a r A t d n a s n o i t c i v n o c , s e c n e t n e s n o s c i t s i t a t S : 2 x i d n e p p A
y t d s e e y e r b n m F a
s e c n e t n e s l a i d o t s u c n o N S E C N E T N E
S D N A S N O I T C I V N O
1 . 2 A E L B C L A A T N I M I R
C L A R E N E
G
3 3 1 0 0 2 2 1 6 0 1 3 6 7 1 3
2 2 2 0 0 7 7 3 6 1 6 5 5 9 6 6 1 0 0 8 2 5 5 2 4 4
r e h t O
3 2 2 5 4 2 7 2 2 3 6 4 3 3 1 4 9 7 3 8 5 8 3 7 9 5 1 7 8 4 7 8 6 9 1 4 2 2 6 5 2 6 7 6 4 1 0 0 9 1 2 9 0 1 8 4 6 0 4 2 9 0 8 1 1 2 1 2 4 4 3 2 3 2 1 1 1 1 2 1 1 1 1 1 1 2 7 2 9 2 2
l a i k r c o o S w
6 9 0 6 2 4 4 2 4 7 8 7 7 7 8 3 6 7 7 9 5 9 4 0 9 7 0 2 2 5 1 1 5 6 2 6 5 5 1 8 7 5 9 0 4 5 3 6 9 6 5 6 1 3 8 4 0 0 2 2 1 1 2 6 6 7 0 3 3 0 9 2 2 7 3 3 4 4 1 2 1 4 3 5 3 3 5 0 9 9 1 1 1 1 1 1 2 3 4 5 6 5 3 2 9 1 2 3
e n i F
0 3 5 3 1 1 6 5 7 2 7 7 4 1 2 5 2 6 7 8 3 6 9 5 3 0 2 3 4 0 7 3 1 0 9 2 0 1 8 1 8 7 9 8 5 3 1 2 6 1 0 7 4 6 0 0 1 4 2 2 7 3 3 1 3 6 3 7 1 6 3 4 8 9 0 3 1 5 7 4 6 1 1 1 7 5 5 9 9 9 1 1 3 4 4 5 3 3 2 2 2 3 4 4 3 3 3 4 2 2 2 2 7 1 9 5 5
l a n e c o i n t i e t d n n e o s C e v r k i t u r c o e b o r r a w o l C l l A
s r a 0 e 1 y 1
m o d e 0 s r e a r 1 ± e f f 6 y o s s o L s r 5 a ± e 2 y
8 5 9 1 4 0 3 6 5 2 1 2 9 9 3 3 3 5 1 3 3 7 0 1 5 6 1 8 7 3 9 6 5 6 2 1 0 7 0 1 3 9 5 7 6 4 7 2 0 2 7 5 6 1 3 0 1 8 4 6 3 5 7 1 8 0 9 7 7 1 7 4 7 2 3 7 0 6 9 6 3 0 4 2 9 3 6 5 5 8 7 6 6 6 8 7 3 4 9 2 9 9 7 6 6 5 5 4 3 5 5 6 2 0 3 1 1 0 1 1 1 6 5 6 2 4 1 3 7 8 2 4 7 2 0 5 3 8 5 2 7 1 4 5 6 5 4 3 2 7 2 1 1 5 5 8 6 4 9 8 9 4 0 0 2 0 2 5 6 5 5 4 3 7 7 5 6 2 0 7 9 1 1 2 0 9 1 2 9 1 7 5 4 4 2 5 9 8 0 8 4 3 7 4 1 5 8 0 9 4 3 0 4 3 4 5 4 1 4 8 2 9 2 0 7 4 4 4 5 8 3 6 9 8 6 4 3 3 3 3 2 1 1 2 1 2 2 1 1 1 1 1 1 1 1 6 8 5 2 3 8 8 0 7 1 7 2 1 4 9 7 4 9 5 1 5 7 3 1 1 6 2 8 5 8 4 7 0 0 7 3 4 3 3 4 9 6 9 0 7 8 1 6 4 5 9 8 5 7 6 2 8 9 3 1 2 5 5 1 5 0 2 4 7 9 8 4 2 6 8 1 9 7 7 9 2 2 8 9 6 1 2 7 3 9 9 0 4 3 0 7 7 9 3 4 0 6 5 0 2 1 2 7 6 0 0 7 1 5 7 9 9 3 9 2 3 3 3 6 4 5 4 4 4 6 9 6 6 5 4 5 4 3 3 4 7 5 3 7 8 0 3 7 3 9 1 5 1 1 0 1 9 1 0 4 9 7 0 8 0 5 5 2 1 0 2 4 0 8 1 6 6 1 1 7 3 3 9 3 5 5 2 7 9 5 4 1 1 1 1 1 2 1 9 9 8 0 3 0 1 5 5 2 7 6 1 1 6 7 1 9 7 3 6 6 6 1 5 6 5 4 5 3 2 2 2 8 8 2 2 8 5 5 3 3 6 9 8 3 1 8 3 2 3 6 0 2 8 8 7 3 0 3 8 6 7 0 2 4 5 9 7 9 3 0 3 8 2 9 4 2 8 1 5 4 9 4 1 0 4 1 7 9 7 9 3 1 4 1 4 3 6 4 3 4 9 6 6 2 4 3 4 6 6 9 2 2 7 4 6 0 2 4 6 9 9 7 5 8 7 3 1 4 2 1 2 5 6 4 3 2 4 3 6 2 8 5 6 3 8 8 9 6 4 1 2 2 2 1 1 1 1 5 6 1 1 8 3 7 7 0 8 7 6 8 9 6 5 5 7 4 3 6 5 3 5 1 2 3 0 7 5 8 6 3 9 2 8 9 3 2 5 4 9 8 9 2 3 6 8 9 7 2 7 3 8 9 4 2 3 5 8 5 2 6 8 3 5 2 0 5 4 9 9 9 4 6 8 8 3 2 6 2 3 5 4 1 2 4 9 7 0 3 1 0 0 1 2 6 1 3 3 8 1 3 4 4 4 8 5 1 8 8 1 2 9 1 7 8 8 6 2 2 2 3 3 3 2 2 2 3 4 2 2 1 1 2 2 1 2 2 4 8 3 3 4
s r 1 a ± e 0 y
2 0 5 0 5 4 5 2 2 8 3 5 3 4 1 8 0 3 0 1 2 5 7 6 3 9 8 9 7 1 2 7 8 0 0 9 6 4 9 4 9 4 2 9 4 3 1 1 8 2 9 5 6 1 4 7 6 8 4 2 5 8 3 3 7 9 2 3 6 0 1 6 2 6 7 3 7 6 8 9 8 5 1 9 0 0 3 4 7 1 5 6 9 9 9 4 2 7 2 4 4 3 3 9 0 8 7 8 6 6 7 9 8 8 6 2 1 1 1 1 2 2 1 1 8 0 1 2
y t h l t a a e n D e p
6 1 7 4 5 0 8 7 2 7 1 7 6 3 4 5 5 2 9 4 1 2 1 6 1 0 6 6 0 2 9 0 6 3 1 1 2 0 7 2 1 2 1 1
s e c l l n A e t n e s
7 3 9 8 1 5 3 0 5 5 5 1 4 4 0 3 7 1 9 2 0 3 3 4 8 1 0 9 4 6 0 0 1 6 0 1 9 5 5 7 0 1 3 3 2 5 8 1 0 3 8 7 9 7 3 9 5 3 3 7 4 4 8 0 7 2 0 7 7 5 3 1 9 9 3 5 7 8 7 2 4 5 0 2 3 7 4 5 9 1 2 0 0 5 0 5 1 2 9 1 9 7 1 1 6 7 8 5 3 9 2 5 8 5 1 6 8 8 8 0 7 7 7 7 0 3 9 8 7 7 8 6 6 6 7 6 6 2 1 1 1 1 1 2 4 1 1 8 2 2 3 5 5 ± ± ± 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 9 7 3 3 3 4 4 4 4 4 4 4 4 4 4 5 5 5 5 5 5 5 3 3 3 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
6 6 6 7 1 8 8 0 7 3 9 2 3 2 2 8 3 2 1 1 4 9 4 2 3
337
6 5 ± 7 3 9 1 r o f s t r u o c l a r e n e g e h t y b d e r e d i s n o c n o i t a g i t s e v n i y r a n i m i l e r p h t i w s e s a c n i t n e m h s i n u p l a n i m i r c f o s e r u s a e m , 9 2 . l , 4 1 . d , s 6 . p o , h c . s 2 9 4 9 . F , F . R R A S S G U : e e c t r h u r o o S f
33 8
STEPHEN G. WHEATCROFT WHEATCRO FT
d e t r a e c h o t l O l a n u
) ’
S E R U G I F V O L G U R
K ` ( 3 5 9 1 ± 1 2 9 1 ,
h t a e d o t s e c n d e t e c n n e s e t l n a e s c i t i m l o o h p w l l y A B ) i i (
S E C R O
Y T I R U C E
T S E V N
I ,
S E C N E T N E
S D N A S N O I T C I V N O
C L A C I T I L O
2 2 9 1 8 9 9 2 6 5 8 5 9 2 8 9 5 4 1 7 7 2 5 9 0 7 0 1 9 6 5 6 0 2 5 0 2 8 1 4 6 6 1 1 1 1 2 1 8 3 3 3 4 2 1 8 2 2 }
i k i o r T s e c n l e a r i e c f e p n o S c
3 4 9 6 6 0 7 8 2 2 6 7 4 5 4 7 9 1 1 8 9 1
0 0 0 0 0 4 }
m U u P i g e G l l O o f C o
7 9 0 3 9 1 7 4 8 2 5 7 4 3 2 1 1
s t s l r a u n u o b C i r T &
1 2 4 0 3 0 3 9 9 1 1 8 4 6 9 8 4 8 2 9 1 8 9 9 2 6 5 8 5 9 2 8 3 8 5 0 6 1 5 3 9 6 6 0 0 5 2 5 5 2 1 7 1 5 4 1 7 7 2 5 9 0 7 0 1 9 7 4 5 7 9 4 5 4 9 3 8 1 2 6 7 1 0 2 1 0 6 5 6 0 2 5 0 2 8 1 4 6 6 1 4 3 0 9 1 2 2 2 2 0 0 2 2 2 1 1 3 8 2 1 8 3 3 3 4 2 1 1 1 9 0 4 2 1 5 2 2 9 4 5 3 3 7 7
0 9 5 7 1 0 7 7 9 7 5 5 6 2 4 5 4 3 6 1 7 6 7 1 4 2 6 4 8 9 2 0 7 5 6 1 1
i k i o r T
F
2 . 2 S E A H T E Y L B B D E T A A T G I
0 0 0 0 6 1 8 4 6 9 8 8 2 5 5 2 1 4 7 1 0 2 1 1 2 2 2 1 1
d e c n e t n e s m o h w y B
s e l c a n i e c r e e p f S n o c m U u P i g G e l l O o f C o s t s l r a u n u o b C i r T & r e h t O e l i x E & s s n p o s i m a r C p
s e . c n N . e t P n M . e s V l a c i t i d l e : o e l p c l c n l n A e l e t t n A n e ) s e i S (
1 9 8 0 3 3 5 1 0 3 8 1 3 0 4 2 9 1 8 3 3 1 4 8 1 1 1 6 4
7 1 0 0 4 7 7 3 4 1 9 8 3 1 1
d e 6 1 2 7 4 7 7 2 2 0 3 2 9 1 8 1 2 4 8 4 1 1 0 3 0 9 6 8 4 9 1 4 c 2 0 4 4 9 7 9 5 3 0 5 6 1 6 2 1 3 9 3 1 8 2 9 7 1 7 5 0 0 8 7 n 6 5 6 3 2 1 9 8 1 3 5 0 3 0 4 9 9 7 0 9 5 5 1 6 5 3 3 2 4 4 0 9 2 6 5 2 e t 9 9 3 5 5 7 9 4 6 5 1 9 8 7 5 3 2 6 7 5 0 6 8 3 8 9 9 8 2 9 n 7 3 1 1 1 2 3 1 1 2 2 2 1 1 4 1 4 2 7 2 1 2 1 1 3 4 2 e s 6 4 3 o h w 4 3 4 6 2 2 7 4 8 0 8 4 7 8 7 8 7 4 7 2 5 2 9 2 9 9 7 0 n 8 2 3 5 6 7 5 0 8 6 6 9 7 5 0 9 6 7 1 1 6 2 9 4 9 3 4 0 o 2 3 4 7 2 0 3 6 5 0 8 8 8 8 3 4 6 9 4 0 6 2 6 8 1 7 6 8 2 2 3 3 9 6 a 0 3 2 5 0 8 8 6 3 4 6 4 5 6 6 1 5 7 8 4 3 7 t 1 1 1 4 5 2 4 4 5 6 9 1 6 5 3 4 4 2 9 4 7 a 1 8 8 7 d o n º 9 3 4 9 7 5 3 4 7 1 6 0 6 4 2 0 9 5 3 7 6 8 9 5 0 7 0 8 0 7 0 3 1 1 3 5 4 3 6 0 6 4 3 5 4 2 8 4 9 5 3 0 9 3 3 1 1 3 9 1 0 4 1 2 1 1 7 4
7 9 8 1 5 2 2 1
7 6 1 7 1 9 3 8 5 8 0 5 4 9 8 8 0 9 8 1 8 7 8 8 0 5 9 1 3 2 9 0 3 9 7 3 4 0 9 2 4 9 0 1 1 8 8 8 1 4 8 2 6 5 5 9 0 7 9 9 7 4 7 9 4 6 1 0 7 6 0 2 3 4 4 4 9 2 8 2 2 2 1 8 6 9 4 2 3 4 5 5 2 9 7 9 1 3 4 1 9 4 1 6 0 6 3 2 2 1 5 1 5 6 7 1 2 4 1 4 3 1 1 1 2 1
7 6 4 4 4 1 5 0 7 6 9 7 2 4 1 9 6 2 3 2 0 0 7 9 7 8 6 9 6 5 5 3 8 2 0 0 1 6 4 2 7 7 3 4 1 1 6 1 6 9 0 1 6 4 8 4 0 7 8 4 4 9 6 1 1 2 2 7 3 1 6 0 8 1 0 7 2 5 2 6 5 8 2 0 2 9 6 7 3 8 7 1 2 0 7 6 6 4 6 4 3 2 4 7 5 1 6 1 2 5 6 8 1 5 4 8 3 6 4 5 3 3 1 6 3 2 1 7 4 1 1 3 5 3 0 5 3 1 2 4 1 1 2 5 6 3 5 3 2 1 1 4 1 4 6 6 1 1 7 7 1 3 3 3 6 3 1 6 8 1 9 6 7 0 9 7 0 1 3 1 2 9 6 2 4 4 7 6 7 2 5 3 5 5 4 6 1 5 4 8 4 0 5 4 1 1 0 6 2 0 0 8 1 8 4 8 5 0 6 4 2 9 9 2 9 7 6 3 1 8 5 2 2 8 4 6 9 9 4 8 4 3 5 6 7 0 8 8 6 6 9 5 5 5 4 1 8 8 3 8 3 1 2 2 4 6 7 2 6 5 4 5 3 8 9 5 9 9 5 4 5 5 8 8 3 6 7 6 2 4 4 9 5 7 4 1 4 2 1 1 2 1 0 7 3 5 8 1 2 0 5 6 6 8 6 7 1 1 7 7 6 5 4 2 3 0 9 1 1 1 1 2 4 2 1 1 6 1 9 2 2 1 2 4 0 3 0 3 9 9 1 1 8 4 6 9 8 4 8 2 9 1 8 9 9 2 6 5 8 5 9 2 8 3 8 5 0 6 1 5 3 9 6 6 0 0 5 2 5 5 2 1 7 1 5 4 1 7 7 2 5 9 0 7 0 1 9 7 4 5 7 9 4 5 4 9 3 8 1 2 6 7 1 0 2 1 0 6 5 6 0 2 5 0 2 8 1 4 6 6 1 4 3 0 1 2 2 2 2 2 2 2 1 1 1 8 3 3 4 2 1 1 1 9 0 4 2 9 0 0 3 8 3 2 1 5 2 2 9 4 5 3 3 7 7
9 3 4 5 5 4 6 7 0 9 6 9 4 9 6 0 5 8 9 6 1 6 1 9 8 4 0 9 5 1 5 0 3 6 5 4 2 0 9 2 9 0 3 5 2 6 9 1 6 9 7 7 6 5 8 0 1 0 4 0 4 9 1 6 2 4 7 0 0 0 9 2 8 0 7 4 9 8 0 7 2 0 6 9 6 9 0 6 6 2 8 8 4 4 4 1 2 2 8 2 1 6 7 8 4 3 0 0 5 6 4 2 5 7 6 3 6 8 0 1 9 8 7 4 0 4 3 1 5 4 8 5 3 3 8 3 5 0 4 8 8 0 1 7 3 1 1 1 2 3 5 0 8 4 3 7 6 7 9 5 6 7 7 2 7 7 2 2 7 7 7 6 5 2 6 8 0 2 1 1 2 2 2 7 5 1 1 1 0 0 1 4 3 1 3 3 2 5 5 5 ± ± ± 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 1 4 9 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4 4 4 5 5 5 5 2 3 3 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
r e b m e t p e S . ± 4 t s 5 u 9 g 1 u A y r , a 1 u l . n o a v J , 5 8 3 n 9 o 1 v ± e 7 3 . h c 9 8 h 2 s 1 , g . u r o p h l , 2 K o r i , t 2 & r 9 v a 9 o M 1 k i i , n k y e s v l i a d h a k M r r a t o g n i e n ) y e n D L e V v d t s M n a e f , h o 2 c e t d 8 2 ± O a e 0 , ( h 8 ’ ) 2 a v . y l o p i s g p t , a u t r ) 5 e K 9 r 9 p r l 1 a e r t e , n n w i e o c h G s k i y o i b M i t ( k n 6 i n e 3 h s 9 c 1 o d n ± t s a 6 i ( 3 2 . 5 9 1 g 9 , g / 1 e 3 2 h c 5 1 h s 9 / i 1 1 ± 1 b 3 d d a 2 t e l 9 a k 1 d e , , o i k i s s s D v o V o k R M ’ i f n o o a g k s a t t p V e e , v D o 2 . s l p a y v i r c e v o i , p k r r S s e t i t s p i 1 s y f e n y n o n e ’ l v v t o e l r s r v t a a s s d P a . u R ) s l e 5 o n e 9 e 9 G o s ` l 1 , o : , v C y c g o n r p y e u o b g b P d a s . e y r e P r b t a e . p s P e n V r o t : p i S t ( e u c e c 7 r l b e 3 u a x o T E 9 1 S
VI CT IM S O F TH E SO V IE T SE CR E T PO L IC E r t 1 e n b 3 e s n m e r o e P c e D
0 0 0 7 3 6 1 7 5 8 4 6 8 4 6 9 9 7 1 0 7 2 0 1 0 0 0 0 8 0 8 6 9 0 2 9 3 9 0 9 3 5 6 0 6 0 7 5 0 7 3 3 4 4 8 3 1 4 5 5 7 5 5 8 8 0 3 3 7 2 9 0 2 8 4 0 5 9 0 6 7 4 0 5 9 3 5 3 8 8 6 6 3 1 7 7 1 6 3 1 2 3 2 9 1 4 0 1 9 6 1 8 0 0 1 1 3 1 2 9 2 2 3 5 7 8 8 9 3 3 5 4 9 6 7 5 8 1 2 4 5 7 7 8 1 1 1 1 1 1 1 1 1 1
d e i D
7 7 5 8 5 6 6 2 5 7 7 7 8 8 4 8 5 9 3 7 4 5 9 9 9 2 9 7 4 0 6 9 7 6 4 4 5 6 0 3 0 8 0 2 1 2 2 3 5 3 5 5 6 9 8 9 9 8 1 6 6 7 7 5 6 8 3 7 6 8 0 5 0 0 6 0 8 6 0 3 8 5 7 5 4 5 0 5 1 6 2 2 2 2 9 5 4 0 4 6 6 4 1 3 2 1 1 1 1 1 2 1 2 2 2 2 2 2 2 2 2
d e t a r e b i L
2 2 2
2 2 2 2
2
2 8 3 2 8 0 9 3 7 5 8 3 1 8 7 1 6 8 2 2 2
? 2 5 4 7 6 2 5 6 8 3 1 0 0 6 8 9 0 9 6 2 0 7 3 4 3 6 2 2 7 3 5 3 5 0 8 4 4 1 6 4 5 0 2 0 5 4 9 6 8 2 5 1 1 7 7 8 1 4 2 2 4 3 0 7 1 9 6 9 3 6 4 9 6 2 6 5 4 1 8 6 4 9 7 0 4 1 6 3 7 2 1 2 0 3 5 3 1 9 6 7 1 5 2 3 0 1 2 3 2 2 3 6 5 3 1 3 1 1 2 1 2 2 3 9 1 2 2 2 2
339
2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2
3 3 9 0 0 0 4 4 0 4 9 9 4 4 5 6 0 3 4 6 2
2 2
0 0 0
r t e e t N h o
) ’
S E R U G I
F
V O K S M E
Z ` (
? 6 1 1 9 8 3 5 5 0 3 6 9 2 5 2 2 6 8 3 9 7 7 5 0 7 5 9 2 1 2 9 6 9 4 9 5 9 3 7 4 9 4 6 7 2 1 5 7 1 1 9 1 3 2 9 9 5 5 9 1 8 6 9 2 3 5 3 9 1 1 1 1 2 2
2 2 2
2
2 2 2 2 2 2 2 2 2
2
2
7 1 3 8 9 7 3 7 4 0 3 7 1 4 5 2 2 2
r o e T h t o
8 3 2 5 6 1 2 4 7 4 0 5 1 8 2 6 3 5 8 9 9 8 3 2 3 5 3 8 1 4 9 0 7 8 6 0 3 9 7 4 2 3 8 7 5 6 4 9 9 3 5 1 7 3 1 0 3 2 5 9 1 2 1 2 6 3 6 6 2 7 7 6 9 2 2 3 1 1 1 1 1
1 6 9 6 5 7 2 5 2 9 9 6 1 8 1 1
r m e o r h t F o
4 2 1 6 8 8 7 9 7 1 4 6 9 3 0 4 9 3 5 0 7 1 8 1 5 9 3 5 0 2 9 3 7 4 7 5 2 3 3 4 3 1 7 3 2 4 2 2 3 1 5 0 8 0 3 8 1 1 1 1 7 7 6 7 0 4 2 2 1 1 1 1
4 5 6 7 6 0 8 7 8 8 5 8 4 5 0 0 0
n i a g t e N
? 0 8 5 2 4 4 5 4 4 8 8 7 3 9 0 7 0 4 7 8 0 0 3 0 2 0 5 9 7 6 3 6 4 4 3 8 6 5 5 7 4 7 0 7 5 4 8 3 4 9 0 8 1 7 2 4 1 7 8 8 9 3 3 0 6 1 2 2 9 2 2 4 6 3 1 1 1 2 1 2 2 2 2 2 4 3 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2
8 4 7 5 4 3 1 9 1 2 0 4 3 3 4 1 2 2
2
y a w a n a R y d a m w e n o r r a u t F n e u r r
0 3 3 4 3 3 3 2 2 2 6 6 2 9 1 3 7 8 3 5 9 9 1 6 3 3 1 9 2 4 8 9 4 7 6 8 7 1 5 8 4 4 3 2 0 3 8 5 8 2 5 1 6 7 2 5 5 3 2 7 3 7 8 8 2 2 1 0 1 6 3 2 2 3 4 2 2 2 1 8 6 5 5 3 1 1 1 1
7 7 5 9 9 7 2 9 3 0 7 8 9 7 7 3
2 8 1 0 9 8 9 8 4 4 9 3 3 9 4 3 3 1 5 5 5 8 9 6 7 3 3 2 8 7 3 5 0 9 9 3 2 4 0 1 7 9 8 4 6 8 8 5 9 0 8 9 2 5 4 7 7 3 9 4 1 1 2 1 1 1 6 5 5 5 2 9 8 6 4 3 1 4 4 3 3 2
9 3 8 3 5 3 1 0 2 8 7 4 5 4 3 2
t e N
3 5 9 1 5 3 9 1 1 4 1 1 3 6 4 8 1 1 4 4 1 2 2 2 4 7 3 4 0 0 9 3 0 6 6 3 9 2 7 6 6 9 1 1 7 9 9 2 9 6 6 7 8 2 1 0 3 7 4 2 2 4 3 3 7 4 1 5 6 5 6 9 2 9 5 3 6 1 5 1 2 3 6 5 2 2 1 3 2 3 2 0 0 8 1 2 2 3 3 4 3 1 2
6 5 6 6 8 1 8 5 3 0 3 7 4 8 3 1 2 2 1 1
s r e r e D p s u h D t t V r m E o K a a L p o I t N c e A D T E r s e s D l D p h a t V v o m i r a r m K N c A o r f
2 0 4 7 6 4 8 5 4 6 9 8 7 9 8 2 9 6 9 0 0 9 8 0 6 4 3 0 7 5 1 3 4 9 3 6 6 3 1 4 0 1 4 6 4 4 3 2 1 7 1 4 6 8 9 7 2 8 6 2 3 2 0 4 0 7 3 0 2 0 4 6 2 3 3 9 8 0 1 8 0 7 7 1 4 4 6 4 5 4 6 9 8 5 0 3 5 5 2 7 1 1 2 2 3 5 5 2 1 1 1 2 2 2 2 2 2
3 4 3 7 4 3 0 4 4 5 5 4 5 1 9 4 5 5 3 4
9 5 5 6 1 7 9 4 3 2 8 7 4 3 2 0 0 7 3 4 8 6 5 8 2 1 9 6 7 5 2 0 4 3 0 0 6 5 9 0 3 2 3 4 7 4 3 9 2 1 4 7 8 6 1 8 6 5 0 5 0 7 7 1 2 8 8 8 6 4 8 9 2 1 3 4 1 7 3 3 0 6 5 1 0 4 9 8 4 1 4 5 7 2 1 6 6 5 0 9 1 1 2 2 3 4 4 2 1 1 1 2 5 5 6 6 3
7 9 5 0 2 4 2 0 7 4 9 1 1 9 3 4 6 8 4 5
m s o r n o f s t i r e p N
7 8 7 6 3 7 2 4 5 6 5 5 4 0 3 7 1 7 6 7 9 9 1 8 1 3 1 1 1 7 0 3 2 3 3 6 9 9 5 4 8 1 1 0 6 6 8 2 1 7 1 0 6 6 9 2 5 5 8 4 4 9 9 0 8 0 7 2 7 9 7 5 8 5 7 0 2 5 1 1 4 3 2 7 8 2 8 0 9 4 0 8 0 5 1 8 9 6 6 8 7 5 3 3 4 3 4 5 7 3 5 7 3 2 2 2 3 5 3
0 8 8 5 9 8 0 3 6 0 2 6 4 2 8 3 2 7 1 4 6
s o n o s T i r p
9 6 6 6 0 2 3 7 7 3 3 7 2 2 1 4 2 5 6 4 6 7 2 1 9 8 1 3 7 9 0 8 3 8 0 4 8 4 3 3 1 9 8 9 7 8 2 5 5 0 3 1 3 7 9 3 8 8 8 9 7 8 3 3 5 4 7 5 6 0 9 0 9 8 0 6 6 1 5 8 1 2 2 4 5 7 5 3 8 4 3 7 9 5 0 1 1 2 1 1 1 1 1
6 4 5 0 1 2 7 7 3 9 3 2 9 3 1 0 2 1 1
s m n o o s r i r F p
7 3 2 9 7 4 7 2 3 8 8 1 2 5 8 5 9 1 9 3 8 6 4 4 0 9 2 1 8 2 2 2 6 4 9 3 3 9 4 5 1 6 4 7 0 9 9 7 5 7 9 1 5 3 4 2 3 2 8 8 5 9 1 6 3 3 4 0 4 5 6 1 1 4 2 8 1 5 4 6 4 0 3 3 0 8 4 4 4 5 2 6 6 2 8 8 7 5 1 1 4 4 4 6 8 3 6 8 5 3 3 3 4 6 4
6 2 3 5 1 1 7 1 0 9 6 9 3 5 9 4 2 9 1 5 7
n o y r t a n u e s n a e r J P 1
0 0 0 0 7 3 6 1 7 5 8 4 6 4 4 6 7 9 7 1 0 7 2 0 0 0 0 0 0 8 0 8 6 9 0 2 9 7 9 0 9 3 5 6 0 6 0 7 0 1 7 3 3 4 4 8 3 1 4 5 5 9 5 5 8 8 0 3 3 7 2 9 9 2 8 4 0 5 9 0 6 7 4 0 5 3 3 5 0 8 8 6 6 3 1 7 7 1 6 3 1 2 3 2 9 1 4 0 1 8 6 1 0 0 0 1 1 3 1 2 1 2 2 3 5 7 8 8 9 3 3 5 4 9 6 7 6 8 1 2 4 5 7 7 1 1 1 1 1 1 1 1 1 1
3 . A T 2 A A D E G N L I T B N U O A C T C
A G A L U G
2
2
2
2 2 2
2
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 3 3 3 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4 4 4 5 5 5 5 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
8 2 3 3 5 5 ± ± ± 7 9 4 3 3 3 9 9 9 1 1 1 l l l a t a t a t o o o T T T
. 4 9 9 1 e n u J ,
w e i v e R l a c i r o t s i H n a c i r e m A , v o k s
m e Z & n r o p s r e t t i R , y t t e G d n a 5 1 ± 4 1 . p p , 9 6 , 1 9 9 1 ,
a y i n a v o d e l s s i e i k s e h c i g o l o i s t o S
, ’ t k e p s a i i k s e h c i g o l o i s t o . s - 3 5 . o l k , i 0 r o 4 t 7 s I 2 ( . g d , a l 1 u . G ` p o , , v 4 o 1 k s 4 9 m . e f Z , . F N R . A V G : : : 3 4 s 5 3 e ± ± c 7 2 r 3 u 3 o 9 9 S 1 1
34 0
STEPHEN G. WHEATCROFT WHEATCRO FT Appendix 3: The unfortunate errors of Norman Davies 63
Norman Davies’ recent book on European History has received great acclaim and is widely available. But in his treatment of Stalinist repression Davies has been greatly misled by Conquest and others. Not only does he appear to believe that Conquest’s high ®gures on the scale of repression have been vindicated, but he offers an account of these ®gures which, I hope, would cause even Conquest to protest. The errors seem to me to be so serious that I feel impelled to list them and correct them. I sincerely hope that future editions of the book will not contain the blemish that the current volume bears. At various places I offer an alternative set of wordings for the incorrect statements. On p. 962 Davies writes: In the 1930s [the main instruments of coercion and terror] were expanded to the point where the th e m anpo an po we r of the th e secu se cu rity ri ty agen ag enci cies es ri val va l led that th at o f the th e Red Re d Ar m y, and an d the th e c amp am p s conta containe ined d up to 10% of the the popul populat atio ion. n. By 1939 1939 the the Gulag Gulag was the the large largesst employ employeer in Europe. Its prisoner-employees, the zeks, who were systematically starved and overworked in arctic conditions, had an average life expectancy of one winter.
Comment: Concerning the population in the camps. If the camps had contained 10% of the population in 1939, this would mean 17 million million people. Conquest Conquest never suggeste suggested d more than 8 million million for the end of 1938, and the currently accepted ®gure is 1.3 million in the camps. 0.3 million in the colonies, about 300 000 in the prisons and about a million in places of special exile, i.e. 1.3 million for the camps alone and 2.9 million for the entire repressive system. It would be better to say th at the camp ca mpss cont co ntain ain ed less le ss th an 1% of t he pop po p ulati ula tion, on, but bu t this th is ®gur ®g uree could co uld rise ri se t o 2.5% 2.5 % if we were to add the populations of prisons, labour colonies and places of special exile. Concerning the level of mortality in the camps, the life expectancy was certainly more th an o ne y ear. ea r. C onq on q uest ue st acce ac cept ptss E kar ka r t’s t’ s clai cl aim m th at oneon e- thir th ir d of the th e new ne w po pu lati la tion on died di ed in its ®rst year. This would imply considerably more than a three year life expectancy and a mortality rate of 33%. However, he points out that life expectancy increased for later years, and he accepts the calculation made by Wiles that average mortality rates were 10% or 10 per p er t ho usan us and. d. Th e ar chiv ch iv al data da ta gi ve a slig sl ig htly ht ly l ow er aver av er age ag e r ate at e of 7% .
On page 964 Davies writes: early in 1939 Stalin and Molotov were signing lists of several thousand named victims each morning, whilst every regional branch of the security police was scooping up far greater quotas of random civilian innocents.
Comment: Th is wo uld ul d b e hig hi g hly hl y e xag xa g gera ge rate ted d f or la te 1 937 93 7 o r ear ly 1 938 93 8 w he n the th e repr re pr e ssio n w a s at its peak. For early 1939, when repression rates were very low, it is totally incorrect.
Fur F ur ther th er on p. 96 4 D av ies ie s w r ites it es:: Fo r m a ny d ecad ec ades, es, o pi nion ni on in the th e o utsi ut side de wor wo r ld w as u nab na b le to c om p reh re h end en d the th e fact fa cts. s. Pr ior io r to t he d ocum oc um enta en tary ry wr itin it in gs o f A l exan ex ande de r So lz heni he nits tsyn yn in the th e 1 9 60s, 60 s, a nd t he p u blic bl icati ati on o f pai p ains nstak tak ing in g re sear sea r ch by a f ew co ur ageo ag eou u s sch olar ol ar s, m o st p eop eo p le i n the th e W est tho th o ught ug ht th at stories of the Terror were much exaggerated. Most Sovietologists sought to minimise it. The Soviet authorities did not admit it until the late 1980s. Stalin, unlike Hitler, did not pay the pri p ri ce of pub pu b lic li c exp ex p osur os uree . Th e to tal ta l tally ta lly of his hi s vict vi ct ims can neve ne verr be exac ex actl tly y calc ca lcul ul ated; ate d; but bu t it is unlikely to be much below 50 millions.
VI CT IM S O F TH E SO V IE T SE CR E T PO L IC E
341
Comment: This Th is is a stran str ange ge and an d highly hig hly i naccu nac curate rate par agr aph. ap h. It im plies pl ies a surpr sur prii singly sin gly posit po sit ivistic ivi stic atti at titu tude de to `the `t he f acts’. act s’. It gives gi ves a total to tally ly wro wr o ng accou acc ount nt of the histo hi story ry of W estern este rn study stu dy of Sovi Sov i et forc fo rced ed labour. The following would be more accurate: `Before Conquest and Solzhenitsyn the disci plin pl in e, especia espe cially lly in Am erica, eri ca, was wa s dom do m inat in ated ed b y the wor wo r k of Dalli Dal lin n and Nicol Ni colaev aevsk sky, y, whi wh i ch even ev en Conquest considers to have given an excessively high evaluation of the scale of the camps and repression. The publication of the painstaking work by a few courageous scholars (Jasny and Timo Ti moshe shenko nko)) who wh o tr ied to establ est ablish ish a m ore r ealist eal istic ic scale was wa s lar gely gel y igno ig nore red. d. And An d attem att empts pts by Wheatcroft Wheatcroft t o revive and extend extend their their arguments did not earn him great great popularity, popularity, although although his arguments arguments were eventuall eventually y vindicated. vindicated. Most Sovietolog Sovietologist istss sought to maximise maximise the scale scale of 64 Terr Te rro o r. The Soviet authorities did not admit it until the late 1980s . Stalin, like Hitler, avoided being being held held respons responsibl iblee for his his action actionss. The total total tally tally of his victims victims can never never be exactl exactly y calculated, but is normally considered to have been about 20 million.
In a footnote 35 Davis states: For Fo r decad de cades, es, many ma ny histo hi storr ians ia ns coun co unted ted Stali Sta lin’s n’s victi vi cti ms in `hu `h u ndred ndr ed s’ or `t housan hou sands’, ds’, whils wh ilst t others, such as Solzhenitsyn, talked of `tens of millions’. Since the collapse of the USSR, the high hi gh est esti mates ma tes have hav e been be en v indic in dicate ated. d. See R. Conqu Co nqu est, The Great Terror. A Re-assessment (London, 1992); also Conquest’s review of the semi-repentant `revisionists’ (J. Arch Getty & R. TL S , February 1994. Manning (eds), Stalinist Terror: New Perspectives ( Cambridge, Cambridge, 1993)), in TLS Yet no precise statistical breakdown has been produced. Studies based on the `demographic gap’ of c.27 million for 1941±45, for example, make no distinction between Soviet citizens killed by the Nazis Na zis and an d those th ose kill ki lled ed by the Sovi So viet et regi re gime me itsel it self. f. No pr oper op er analys ana lysis is of losses lo sses in the th e USS R by nati nationa onalit lity y has has bee been forth forthco comin ming. g. See See Norma Norman n Davie Davies, s, `Nei `Neithe therr Twenty Twenty Milli Million on,, nor Russia Ru ssians, ns, nor no r W ar Death De ath s’, Independent , 29 December 1987; also M. Ellman. `On Sources: A Note’, Soviet Studies, 44, 5, 1992, pp. 913±915.
Comment: Th e fo llo ll o wing wi ng i s a m or e accu ac curr ate v er sion si on o f even ev en ts: For Fo r decad de cades, es, many ma ny histo hi storr ians ia ns coun co unted ted Stali Sta li n’s victi vi cti ms in `tens `t ens of m illi il lion ons’, s’, which wh ich was wa s a ®gu re supp suppo orte rted by Solz Solzhe heni nittsyn. Sinc Sincee the the coll collap apsse of the the USSR, SSR, the the low lower esti estima mate tess of the scale of the camps have been vindicated. The arguments about excess mortality are far Great Terror Terror:: A Re-ass Re-assess essme ment nt more complex complex than than normall normally y believ believed ed.. R. Conques Conquest, t, The Great (Londo (London, n, 1992) does does not not real reallly get to grips grips with with the new new data data and and cont continu inues es to pres presen ent t an exa exagg gger eraated ted pic pictur ture of the the repr repres esssion ion. The The vie view of the `re `revis vision ionists’ ts’ has has bee been largely substantiated (J. Arch Getty & R. T. Manning (eds), Stalinist Terror: New Perspectives 65 TL S and Independent , have (Cambridge, (Cambridge, 1993)). The Th e popu po pular lar press, pr ess, even ev en TLS an d The Independent have contain contained ed erroneous journalistic articles that should not be cited in respectable academic articles.
On p. 963 Davies presents a Capsule on Vorkuta that includes the paragraph: `Over the years, more human beings perished there [in Vorkuta] than at Auschwitz; and they died slowly, in despair. But few history books remember them’. Ref Re f ere er e nce nc e to Pau Pa u l H oll ol l ande an der, r, ` Sovi So viet et Ter Te r r or, or , A m eri er i can am n esi a’, National Review , 2 May 1944, pp. 28±39.
Comment: Th e fo llo ll o wing wi ng i s a m or e accu ac curr ate st atem at emen ent: t:
34 2
STEPHEN G. WHEATCROFT WHEATCRO FT TA T A B L E A3 .1 O RMAN D AVIES ’ E STIMATE N ORMAN
OF
ON - W ORLD O RLD W AR AR II D E A T H S E XCESS N ON (M ILLIONS )
I N USSR, USSR,
C i v il w a r a n d V o l g a f a m i n e P o l i t i c a l r e p r e ss i o n o f t h e 1 9 2 0 s Forc Fo rc ed c oll ec tivi sat ion and an d de kula ku la kisa ti on af ter 192 19 2 9 Ukra Uk ra in ian ia n terr te rror or -fa -f a m in e, 19 32 ±33 ±3 3 G r e at t e r r o r ( 1 9 34 ± 39 ) an d p u rg es D e p o r t a t io n t o t he G ul a g , to 1 93 7 S ho o t i n g s a n d r a n d om e x e c u t io n s, 1 93 7 ±3 9 D e p o r t a t io n s f ro m E . P o la nd , B a l t ic S t a t e s a n d R o m a n i a , 19 39 ± 4 0 Fore Fo re ign ig n PO W s: Pol Po l es, Fi nns, nn s, Ge rm an s, Ro m ania an ia ns, Jap an ese D e p o r t a t io n s t o G u l a g , 19 39 ± 45 D e p o r t a t io n o f n a t io n al i t i e s: Vo lg a G e r m a n s, C he ch e n s et c. P os ost -w -w a r sc re re en en in in g o f re pa pa tr tr ia iat es es an d i nh nha bi bi ta ta nt nt s o f e xx- o occ cu cu pi pi ed ed t er er ri ri to tor y G r o ss t o t a l
1917±1953 1917±1953
M in .
M a x.
3. 0 0. 05 10.0 10 .0 6.0
5 .0 14.0 14 .0 7.0 1 .0
1 0. 0
5. 0
1 .0 2 .0 1.0 7 .0 1 .0 6 .0 .0
c .5 4
several of these these categorie categoriess overlap; overlap; for politic political al repres repressions sions in the 1920s I have i nterpret nterpreted ed Notes: several `tens of thousands’ as 50 000. Sourc So urc e : Norman Davies, Europe: A History (Oxford, 1996), p. 1329.
`Over the years there was great human suffering in Vorkuta, but it cannot be compared with the th e sca le of A u sch w it z’. z’ . Ref Re f ere er e nce nc e to S. G . W heat he atcr cr oft, of t, ` Th e Sc ale and an d N atur at uree of Ger Ge r m an an d So v iet Rep r essio ess io n and Mass Killings, 1930±45’, Europe-Asia Studies , 48, 8, 1996, pp. 1319±1353.
And ®nally (see Table A3.1), on p. 1329, in an appendix on Europe’s Estimated Death Toll, 1914±45, there is Table 5, Categories of people killed in Soviet Russia and the Soviet Union 1917±1953 (excluding war losses 1939±45). For sources we are to ld that th at this th is is ` af ter te r R. M e dve dv e de v a nd R. Co nque nq ue st’ st ’ . Comment: Th is cann ca nno o t be t aken ak en ser se r iou io u sly. sly . Th e o nly nl y re deem de emin ing g f e atu re o f this th is list li st is t he adm ad m issio is sion n th at `several of these categories overlap’. It is rather unfair of Davies to attribute such ®gures to R. M edv ed v edev ed ev and an d R. Con Co n que qu e st. I n the th e tabl ta blee Da vies vi es l ists ist s 29 to 3 5 m illi il li on e xcess xc ess d eat hs in 66 the th e 19 20 s and an d 19 3 0s. M edv ed v edev ed ev m ade ad e est imat im ates es o f 1 0 or 12 1 million, and Conquest 67 favo favoure ured d a ®gure ®gure of 20 milli million on.. Detaile Detailed d demographi demographicc estima estimates tes of excess excess mortalit mortality y 68 between the 1926 and 1939 censuses tend to indicate levels of 10 million. Calculations of excess mortality are extremely dif®cult and totally depend on what levels of mortality are taken as normal. The USSR was undergoing a massive and extremely rapid demographic transition at this time. The secular mortality rates were generally falling at a very rapid rate during this period. This makes it even more dif®cult to assign a normal level 69 of mortality. 1
Robe Ro bert rt Conq Co nques uest, t, `Vict `V ictim imss of Stalin Sta linism ism : A Comm Co mm ent’, en t’, Europe-Asia Studies, 49, 7, November 1997, pp. 1317±1319, in response response to Stephen Stephen W heatcrof heatcroft, t, `The Scale and Nature of G erman and Soviet Soviet Studies , 48, 8, December 1996, pp. 1319±1353. Repr Re press ession ion and M ass Killin Ki llin gs, 1930 19 30±45 ±45 ’, Europe-Asia Studies Un fortun for tun ately ate ly I only on ly becam bec am e aware aw are of this thi s articl art iclee in early ear ly Augu Au gust st 1998 19 98.. 2 R. Conqu Co nqu est, `Fo rced rce d Labo La bour ur Statis Sta tistic tics: s: Som So m e Com Co m me nts’, nts ’, Soviet Studies , 34, 3, 1982, p. 438,
VI CT IM S O F TH E SO V IE T SE CR E T PO L IC E
343
in response to my article `On Assessing the Size of Forced Concentration Camp Labour in the Soviet Un ion, ion , 1929± 19 29± 56’, 56 ’, Soviet Studies , 33, 2, 1981, pp. 265±295. 3 The Th e argum arg um ent w ould ou ld hav e justi® jus ti®ed ed a con clusion clu sion of no mo more re than tha n three thr ee to fou r milli mi llion on,, but bu t at a moment of excessive caution I wrote four to ®ve million; see ibid. , p. 286. 4 See the recent work of Norman Davies, which will be discussed in Appendix 3. 5 Conquest Conquest demonstrate demonstratess a rather rather simplistic simplistic as well as partisa partisan n view on what he expects expects from from the data. No one should expect the data to be perfectly compatible. In fact one should be suspicious if they were. Below I w ill argue that the scale of the incompatibil incompatibility ity is far, far less less than Conquest Conquest claims, claims, and that th at this thi s sm small all level lev el of under un dersta stand ndab able le incom inc om patibi pat ibilit lity y shou sh ould ld be tak en as groun gr oun ds for fo r even ev en gre ater ate r faith fai th in these data, and as evidence that they have not been doctored or prepared especially for us. 6 Great Terro Terrorr (Harmon R. Conq Co nques uest, t, The Great (Harmonds dswort worth, h, 1968) 1968),, Pref Preface ace,, p. 14., 14., `But `But the the most most important consideration remains the mere extent of the Terror’. 7 See S. G. Wheat Wheatcr croft oft,, `Popul `Populat atio ion n dynami dynamics cs and facto factors rs affe affect cting ing them them in the the USSR’, USSR’, Birmingham University, CREES, Soviet Industrialization Project Series, SIPS, no. 1. 8 D. J. Dallin & B. I. Nicolaevsky, Forced Labour in Soviet Russia (London, Hollis and Carter, 1948), pp. 52±86. 9 N. S. Timashe Timasheff, ff, `The Post-wa Post-warr popul populati ation on of the Soviet Soviet Union’, The American Journal of Soci S ocio o logy lo gy , 54, 1948, pp. 148±155. 10 Journal al of Polit Politic ical al N. Jasn Jasny, y, `Labou `Labourr and outpu outputt in Sovie Soviett Conce Concentr ntrat ation ion Camps Camps’, ’, The Journ Economy , 59, 1952, pp. 415±416, and `Comments’, 60, 1952, p. 340. 11 I ®nd it bizarre that Conquest is described by Norman Davies as `courageous’ even though he is providing ®gures that supported the popularly accepted Cold War stereotype favoured at the time. Th ose w ho are coura co urage geous ous are tho se w ho act again ag ainst st the popul po pular ar stereo ste reoty types pes of their the ir tim e, e.g . Jasny Ja sny in America in 1952 and Danilov in Moscow in 1989. 12 A. Bergson, Bergson, The Real National Income of Soviet Russia since 1928 (Harvard UP, 1961), p. 433. 13 S. Swianiewicz, Forced Labour and Economic Development: An enquiry into the experience of Sovi S oviet et Indu In dustr stria ializ lizati ation on (Oxford, 1965), p. 38 and Appendix A, pp. 699±713. 14 Conquest, The Great Terror , p. 708. He estimated ®ve million in jail or camp in January 1937 and eight million by late 1938. 15 I will explain this point in more detail below. 16 Great Terro Terrorr , pp. 699±702. Conquest, The Great 699±702. The Casual Casualtty Figur Figures es Appendi Appendix x was remov removed ed from from the new edition edition in 1990, 1990, but, but, as we shall shall see see below, below, Conque Conquest st conti continue nuess to adhere adhere to these these conclusions. 17 Conquest, ibid., p. 705. 18 Ibid. , p. 7 10. `Taking the conservativ conservativee ®gures ®gures of an average over the period 1936±50 inclusive of 8 million population of the camps and a 10% death rate per annum, we get a total casualty ®gure of 12 million dead. 19 S. Rose®e Rose®eld lde, e, `An Assessme Assessment nt of the Sources Sources and and Uses Uses of Gulag Gulag Forced Forced Labour, 1929±56 1929±56’, ’, Sovi S oviet et S tudies tud ies , 33, 1, 1981, pp. 51±87. 20 R. W . D avies avi es & S. G. W hea tcroft tcr oft,, `Rose `R ose®el ®elde de’s ’s K liukva liu kva ’, Slavic Review , December 1980, pp. 593±602, and S. G. Wheatcroft, `On Assessing the Size of Forced Concentration Camp Labour in the Soviet Union, 1929±56’, Soviet Studies , 33, 2, 1981, pp. 265±295. 21 Wheatcroft, Soviet Studies , 33, 2, 1981; 35, 2, 1983; and 48, 8, 1996. 22 R. Conqu Co nqu est, `Fo rced rce d Labo La bour ur Statis Sta tistic tics; s; Som So m e Com Co m me nts’, nts ’, Soviet Studies , 34, 3, 1982, p. 438. 23 One of the sour source cess that that Conq Conque uesst accu accuse sed d me of not not chec checki kin ng was was a calc calcul ulat atio ion n by S. Voronitsyn in a Radio Liberty Research Report to the effect effect that the disenfranc disenfranchise hised d populatio population n in 1938 was 17.6 million and not the 4.7 million that I claimed. However, on closer investigation Keith Bush from Radio Liberty wrote to me to apologise for Voronitsyn’s error in comparing the number of voters in 1939 in the pre-1939 territory of the USSR with the total eligible to vote in 1939 in the post po st-19 1945 45 territ ter ritor ory y of the th e Sovie So viett Un ion. ion . W hen he n this thi s w as eve ntuall ntu ally y point po inted ed out ou t to Conq Co nque uest, st, he wrot wr otee to the th e editor edi torss of this thi s journa jou rna l w ith an inter in terest esting ing retrac ret rac tion , w hich hic h includ inc lud ed the claim cla im tha t the differ dif feren ence ce between 17.7 million and 4.7 million did not materially affect his arguments. See Robert Conquest, `Letter to the Editor’, Soviet Studies , 35, 1, 1983, pp. 133±134. 24 At the height of this discussion I was able to prevail on David Ransel, the editor of Slavic Review , to commissi commission on two demogra demographe phers rs to look into into this questio question. n. The debate debate sorely sorely needed needed a profe pr ofessi ssion onal al input. inp ut. W hen he n they the y subse su bsequ quentl ently y arg ued tha t exc ess m ortali or tality ty w as a com co m ple x phe nom en on highl hig hly y depend dep end ent upon upo n what wha t w as consid con sidered ered `norm `no rmal’ al’ their the ir views vie ws w ere dism issed iss ed by Conque Con quest, st, who seemed to doubt that demographers had anything to add to this debate. See Slavic Review , Fall 1985. 25 See the forthcoming publication by Robert Johnson, based on the 1996 Toronto Conference on Stalinist demographic data. 26 See Zemskov in conversation with S. Cholak, in Argumenty i fakty , 1989, 45, p. 7.
34 4
STEPHEN G. WHEATCROFT WHEATCRO FT 27
S. G. Wheatcroft, `More light on the scale of repression and excess mortality in the Soviet Union in the 1930s’, Soviet Studies , 42, 2, 1990. Alec Nove also wrote about these materials in Soviet Studies , 42, 2 and 42, 4, 1990. 28 Perhaps the worst example of this was in his interview with K. van den Heuvel, `Testimonies of a non-witness’, Moscow News, 26 March 1989. 29 See Danilov, `Letter to the Editor’, Voprosy istorii , 1988, 3, pp. 116±121 and Conquest’s reply istorii and the attack on Danilov in Moscow News. in Voprosy istorii 30 R. Conqu Con quest, est, The Great Terror: A Reassessment (1990), p. 311. 31 Ibid. , p. 486. 32 R. C onqu on ques est, t, `Exc `E xces esss Deat De aths hs and an d Ca mp N um be rs: Som So m e Co mm ents’ en ts’,, Soviet Studies , 43, 5, 1991, p. 951. 33 See V. N. Zemskov, `Ob uchete spetskontingenta NKVD vo vsesoyuznykh perepisyakh nasetsiologicheskiee issledovaniya issledovaniya, 1991, 2, pp. 74±75. leniya 1937 i 1939gg.’, So tsiologicheski 34 R. Conqu Con quest, est, Letter Let ter to the editor, edit or, American Historical Review , June 1994, p. 1039. 35 Getty & Rittersporn, `Letter to the Editor’, American Historical Review , June 1994. Of course, sever several al of the named named ®gures ®gures were were not post post-S -Sovi oviet et of®ci of®cial als, s, but with with that minor minor quali® quali®ca cati tion on,, this this statement seems to me to be valid. 36 Norman Davies, Europe: A History (Oxford, 1996), republished as a best-seller by Pimlico in 1997. See Appendix. 37 These Th ese w ere publish pub lish ed ®rst ®rs t in 1989 198 9 in the m ass circula circ ulatio tion n new spaper spa per Argumenty i fakty , and from 1991 in the sociological journal Sotsiologicheskie issledovaniya , and more recently, in English in a joint publication with Arch Getty & Gabor Rittersporn, `Victims of the Soviet Penal System in the American Historic Historical al Review Review , Pre-wa Pre-warr Years Years:: A ®rst ®rst appr approac oach h on the Basi Basiss of Archi Archival val Evid Evidenc ence’, e’, American October 1993, pp. 1017±1048. 38 When Zemskov ®rst cited these materials in his articles in 1989±91 they had still not yet been formally declassi®ed and so Zemskov was forced to adhere to the curious rules of the time that did not allow allow archiva archivall citati citation on to be given given to un-decla un-declass ssi® i®ed ed materials materials when publis published. hed. Subsequ Subsequentl ently y these these materia materials ls have have been been declass declassii®ed and his 1996 article, article, V. N. Zemskov, Zemskov, `Zakl `Zaklyuche yuchennye nnye v 30-e gody Sotsiologicheskiee issledovaniya issledovaniya, 1996 (demogra®cheskii (demogra®cheskii aspekt)’, aspekt)’, Sotsiologicheski 1996,, 7, pp. pp. 3±14 3±14 give givess full ull arch archiv ival al references. 39 This Th is point poi nt will wi ll be discus dis cussed sed furthe fur therr below be low.. 40 The Th e con vention ven tion for fo r mo most st of the 1930s 193 0s was to includ inc ludee labour lab our colo nies (ITK ) w ith prisons pris ons in the category `places of detention’, but to exclude the labour camps (ITL). 41 For Fo r M oscow, osc ow, Rasstrel’nye spiski , vyp. 1, `Donskoe kladbishche, 1934±1940’ (Moscow, 1993), vyp. 2, `Vagan’kovskoe kladbishche, 1926±1936’ (Moscow, 1995), and for Leningrad, Leningradskii Martirolog, 1937±1938, vol. 1, August±September 1937 (St Petersburg, 1995), vol. 2 October 1937 (St Petersburg, 1996). 42 Given our explanation of how the series of data were calculated at the local camp level, it is clear that tha t the Gulag Gu lag author aut horitie itie s w ould oul d have hav e needed nee ded to do som ething ethi ng to remov rem ovee the double dou ble countin cou nting g to arrive arri ve at a net in¯ow ®gure. Their tables clearly have not done this. This is not conscious distortion, simply statistical practicability. 43 CR (or K-R in Russian) refers to Counter-Revolutionary. This was the description of crimes charged under article 58 of the 1926 Criminal Code. 44 I.e. all charges in the criminal code apart from article 58. 45 Conqu Conques estt stat states es 25 700 Poles Poles as being being shot shot.. This This ®gure ®gure actu actuall ally y refer referss to the 14 700 Polis Polish h of®cer of®cerss and other otherss in POW camps camps and and the the 11 000 CRs held held in Wester Western n Ukrain Ukrainee and and Belor Beloruss ussia ia,, whom whom Beri Beriaa aske asked d the Poli Politb tburo uro for for perm permis issi sion on to try try in an extr extraa-or ordi dina nary ry manne mannerr and and shoo shoot. t. The Th e Politbu Poli tburo ro did grant gra nt permis per mission sion for this to be done, don e, but as Shelep She lepin in later lat er inform inf ormed ed Khrush Khr ushche chev, v, 21 857 were actually actually shot. shot. All the document documentss concern concerning ing this matter matter have have been been reprodu reproduced ced by the the Katyn:: Documen Documents ts of Genoci Genocide de:: Documen Documents ts and olish olish Acade Academy my of Scienc Sciences es in W. Mater Materski ski (ed.), (ed.), Katyn mater materiials als from from the Sovie Soviett arch archiives ves turned urned over over to Poland Poland on Octo Octobe berr 14, 14, 1992 1992 (Warsaw, (Warsaw, 1993), 1993), pp. pp . 11± 25 and 26±29 26± 29.. 46 Wheatcroft, `The scale and nature ¼’, p. 1331. 47 Krivoshein, Grif Sekretnosti Snyat , p. 140. 48 See `Les Annees 30: Nouvelles directions de la recherche’, Cahiers du Monde russe , 39, 1±2, 1998, pp. 197±208. 49 Khlevnyuk, p. 205. 50 Wheatcroft, `The Scale and Nature ¼’, p. 1351, footnote 37. 51 See ibid ., . , pp. pp. 1327, 1327, foot footno note te 19. 19. Let me immedi immediat atel ely y say say that that the the redu reduct ction ion of the the Ausc Auschwi hwittz mortality ®gures does not at all invalidate the overall ®gures.
VI CT IM S O F TH E SO V IE T SE CR E T PO L IC E 52
345
See the Getty & Rittersporn, `Letter to the Editor’. See Edwin Bacon, The Gulag at War: Stalin’s Forced Labour System in the Light of the Archives (London, Macmillan, 1994), pp. 36±38. 54 The case of the late Dmitrii Dmitri i Volkogonov is perhaps most instructive on this matter. In his latest posthumously posthumous ly published book, D. Volkogonov, Volkogonov , The Rise and Fall of the Soviet Empire (Harper-Collins, 1998), Volkogonov had ample opportunity to repeat these ®gures, if he really thought them correct. But he failed to do so. On p. 153 he cited the Andreev, Darsky & Kharkova results that between 1929 and 1953 `Stalin `Stalin depriv deprived ed 21.5 21.5 milli million on Soviet Soviet citiz citizens ens of their their lives lives’. ’. But But he cited cited Kruglo Kruglov’s v’s report report to Stali Stalin n of 2 188 355 prisoners in ITL and ITK in March 1947 (p. 150), and appears to cite approvingly from the Kruglov 1953 report on p. 185, the special board of the NKVD which had condemned 442 531 people to death and long-term imprisonment, for which he provides an archival reference (APRF, f 3, op. 24, d. 435, l. 41). Why would Volkogonov cite this ®gure in a serious work if he really believed that over seven million had been executed overall, and that 1.75 million had been executed on political charges in 1937 alone? 55 Voennyee kadr kadry Sovet Sovetsk skog ogo o gosuda gosudars rstv tva a v Veli Veliko koii Otec Oteches hestv tven enno noii Voine, Voine, 194 1941±19 1±1945g 45gg. g., See Voenny Tsifrovo Tsifr ovoii sbornik sborn ik (Moscow, (Mosc ow, Voenizdat, Voen izdat, 1963), 1963) , p. 11. 56 Conquest, The Great Terror , pp. 228, 485. 57 G. A. Kumanev, Pravda, 22 July July 1989 1989,, cite cited d here here from A. T. Ukol Ukolov ov,, Deput Deputy y Pres Presid iden entt of the the Military Collegium of the Supreme Court, and V. I. Ivkin, Humanitarian Academy of the Armed Forces of the Russian Federation, `O masshtab repressii v krasnoi armii v predvoennye gody’, Voenno-istoricheskii zhurna zhu rnal l , 1993, 1, p. 56. 58 These Shchadenko Shchade nko report repor t ®gures ®gur es were ®rst published publishe d in Izvestiya TsK KPSS , 1990, 1, p. 188, and have been discussed discu ssed widely in the West by Roger R. Reese, Rees e, `The `Th e Impact Impa ct of the Great Grea t Purge on the Red Army: Wrestling with hard numbers’, The Soviet and Post-Soviet Review , 19, 1±3, 1992, pp. 71±90, and `The Red Army and the Great Purge’, in J. Arch Getty & Roberta T. Manning (eds), Stalinist Terror: New Perspectives (Cambridge (Cambridge,, 1993), 1993), pp. 198±214, 198±214, and in Russia Russia by Ukolov & Ivkin, Ivkin, Voenno-istoricheskii zhurna zhu rnal l , 1993, 1, pp. 56±59. 59 See R. W. Davi Davies es,, M. Tau Tauger ger & S. G. Whea Wheatcro tcrofft, `Sta `Stali lin, n, Grain Grain Stoc Stocks ks and and the the Famin Faminee of 1932±1933’, Slavic Review , Fall 1995, pp. 642±57. Danilov is now himself aware of this error. But since he never put this preliminar prelim inary y ®nding into print he has nothing nothin g to withdraw. withdr aw. 60 See S. G. Wheatcroft & R. W. Davies, `The agricultural crisis of 1931±3’, forthcoming (in Russian) in Otechestvennaya istoriya , and in English in S. G. Wheatcroft (ed.), Challenging the Traditional Views of Russian History , forthcoming (Macmillan). 61 See Davies, Tauger & Wheatcroft, p. 642. 62 As I was completing this article my attention was drawn to the ®rst of a series of detailed volumes on different aspects of the Soviet repression produced by Arsenii Roginsky and his team in Memorial, Repressii protiv Polyakov Polyako v i Pol’skikh P ol’skikh grazhdan. grazhd an. Istoricheskie Sborniki Sborn iki ªMemor ialaº , vyp. 1 (Moscow, 1997). Although this volume only concerns repression against Poles in 1937±38 it indicates the wealth of data which the Roginsky team has in its possession. Several sentences are devoted to placing the arrests of the Poles in comparative comparative perspec p erspective tive with all arrests at this t ime. These sentences indicate that Roginsky and Petrov accept accept the general level of accuracy of the of®cial arrest ®gures, which are comparable with their detailed data. See N. V. Petrov & A. B. Roginskii, ` ªPol’skaya operatsiyaº NKVD 1937±1938gg.’, Repressii protiv Polyakov i Pol’skikh grazhdan, pp. 32±33. We look forward to seeing more of their material. 63 Davies, Europe: A History . It has generally been given rave reviews. According to the TLS , `No history histo ry of Europe Euro pe in the English Engl ish Language Langua ge has been so even-handed even- handed in its treatment treatm ent of East and West’. The Sun Sunday day Times Times claime Felipe Felip e Fernandez-Ar Ferna ndez-Armesto mesto in The claimed d that that,, `It brims brims with with lear learni ning ng,, crac crackl kles es with with Independent t common common sens sense, e, coru corusc scat ates es with with wit and and abou abound ndss in good good judg judgeme ement nt’. ’. Jan Jan Moris Moris in The Independen described described it as `A noble monument of scholarship’ scholarship’ and claimed that `There are superb assessment assessmentss of vastly daunting subjects’. Only in distant Australia did an occasional discerning academic challenge some of his judgements. judg ements. See Peter McPhee, McPh ee, `Europe `Euro pe on the brink’, brink’ , The Age (Melbourne), 5 April 1997. 64 Sentences in italics are sentences in the original text which are correct. 65 Far from fro m being `semi-repe `sem i-repentant’, ntant’, one of their contributor contr ibutor s found their introductio intro duction n a little excessive excess ive on the triump triumphal halism. ism. See introducti introduction on p. 13, footnote footnote 23, `The archives archives thus thus show show that that Wheatcr Wheatcroft’ oft’ss previously prev iously published publishe d estimates estima tes of 4 to 5 millio n prisoners prisoner s maximum maxi mum were amazingly amaz ingly accurate’. accurate’ . 66 See Moscow News, 27 November 1988, p. 8±9, Argumenty i fakty , 1989, 5, pp. 5±6. 67 See above. 68 See estim stimat ates es of Lori Lorimer mer,, Maks Maksud udo ov and and Andre Andreev ev,, Dars Darsky ky & Khar Kharko kova va as summa summarrise ised in S. G. Wheatcroft & R. W. Davies, `Population’, in R. W. Davies, Mark Harrison & S. G. Wheatcroft (eds), The Economic Transformation of the Soviet Union, 1923±1945 (Cambridge), pp. 72±77. 69 For a further furthe r discussion discuss ion of this and an attempt attemp t to place the Soviet Sovie t mortality morta lity crisis in the perspective persp ective of long-term secular mortality decline see S. G. Wheatcroft `The Great Leap Upwards’, Slavic Review , forthcoming, 1999. 53