Heisenberg, Goudsmit and the German Atomic BombMark Walker Citation: Phys. Today 43(1), 52 (1990); doi: 10.1063/1.881237 View online: http://dx.doi.org/10.1063/1.881237 View Table of Contents: http://www.physicstoday.org/resource/1/PHTOAD/v43/i1 Published by the AIP Publishing LLC. Additional resources for Physics TodayHomepage: http://www.physicstoday.org/ Information: http://www.physicstoday.org/about_us Daily Edition: http://www.physicstoday.org/daily_edition

Downloaded 21 Sep 2013 to 128.103.149.52. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://www.physicstoday.org/about_us/terms

HEISENBERG, GOUDSMIT ANDTHE GERMAN ATOMIC BOMB

Contrary to accounts based on Heisenberg's claims, the Germanfission research effort in World War II was indeed a nuclearweapons program, and contrary to Goudsmit's interpretations,the German team knew what it was doing.

Mark Walker

The question of whether German scientists would havebeen willing to make atomic bombs for Adolf Hitler has ex-cited persistent interest. Just why this is so is a topic Ihave explored elsewhere.1 Here, I contend that the rootsof the controversy about the role of the German scientistsare to be found mainly in the period immediately after thewar, not in the war itself.

When I set out several years ago to write a doctoraldissertation on Germany's wartime nuclear program, Ifound that the secondary literature was confusing, contra-dictory and largely undocumented. Almost all the litera-ture turned out to derive either from a polemic interpreta-tion of the German war work set forth by SamuelGoudsmit or from an apologetic interpretation profferedby Werner Heisenberg. In particular, the well-knownbook by Robert Jungk, Brighter than a Thousand Suns(Harcourt, Brace, New York, 1958), did much to amplifyHeisenberg's claim that the German scientists had con-spired to deny Hitler a bomb. While this conspiracytheory was rebutted in David Irving's German AtomicBomb (Simon and Schuster, New York, 1968), Irvingperpetuated Heisenberg's false claim that an erroneouscalculation by Walter Bothe had seriously retarded theGerman effort. Irving's book has influenced books onrelated but broader topics, and even when such books havemet high scholarly standards, they have tended to followin the footsteps of the previous secondary literature.2

This article focuses on the illuminating and troublingdialogue that took place after the war between twocolleagues and former friends, Goudsmit and Heisen-

g—the debate that did so much to shape subsequentliterature.

Mark Walker is a member of the history department atUnion College, in Schenectady, New York. This article isbased on his book German National Socialism and the Questfor Nuclear Power (Cambridge U. P., New York, 1989).

Goudsmit, who had discovered electron spin withGeorge Uhlenbeck (see PHYSICS TODAY, December, page 34)and who for many years was editor in chief of TheAmerican Physical Society, was one of the most influentialphysicists in America after World War II. But he isperhaps best known outside of the physics community forhis work in scientific intelligence. During the last fewyears of the war, Goudsmit served in Europe as a reserveofficer and the ranking scientific member of the AlsosMission, an extraordinary intelligence-gathering unit ofthe American nuclear weapons project. Goudsmit and hiscolleagues hunted down the German scientists who hadbeen involved in applied nuclear fission and isotopeseparation research, in the process seizing scientificreports and materials, destroying experimental apparatusand arresting physicists and chemists. Goudsmit's heroicaccount of his adventures was presented in his book Alsosin 1947 (Tomash, Los Angeles; second edition, AmericanInstitute of Physics, 1988).

In many respects, by the time Goudsmit received hisdischarge from the army, he was an embittered man. Hismother and father had died at Auschwitz. The loss of hisparents and the horrific legacy of National SocialistGermany were blows that he would feel keenly the rest ofhis life. Understandably, Goudsmit no longer was com-pletely objective when it came to Germany, Germanscience or German scientists. Immediately after the war,he advocated a sink-or-swim policy for German science.But he eventually came to soften and qualify his stance to-ward German science, and he did so partly as a result of hisemotional public and private debate with Heisenberg.

Heisenberg, already a world-famous physicist whenthe war began, was an ambitious man and a Germanpatriot who, after being badly bruised in the 1930s byattacks by the physicist Johannes Stark, an early support-er of Hitler, had had to solicit the assistance of the SS to de-fend his position. After the war Heisenberg wo ? eager tobe recognized as a leader of Germany's physics communi-ty.3 In his exchanges with Goudsmit, he sought to portray

5 2 PHYSICS TODAY JANUARY 1990Downloaded 21 Sep 2013 to 128.103.149.52. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://www.physicstoday.org/about_us/terms

Werner Heisenberg, an ardent Germannationalist but not a National Socialist, soughtto dissociate himself and other leadingphysicists in postwar Germany from Naziscience by portraying physics administrators inthe Third Reich as incompetent.

Samuel Goudsmit (left), amember of a militaryintelligence force during thewar and a leader of thephysics community in postwarAmerica, clung tenaciously tothe mistaken idea that theGermans thought an atomicbomb would consist of anuclear reactor in whichthermal neutron reactionswent out of control.

his wartime responsibilities as greater than they actuallywere without incurring the moral onus of having tried tobuild a superbomb for Hitler. At the same time, he tried todissociate himself and most German physicists from thewartime leaders of German fission research by insinuat-ing, without coming right out and saying so, that theadministrators had been contaminated with Nazism, sothat they would be lumped together with the proponents ofdeutsche Physik—the antirelativistic, supposedly moreGermanic theories expounded by physicists like PhilippLenard and Stark (see the article by Fritz Stern in PHYSICSTODAY, February 1986, page 40).

A common attribute of Heisenberg's apologetic ac-count and Goudsmit's polemic was, ironically, a distinctexaggeration of Heisenberg's importance. Heisenbergappeared as though he had controlled and dominated theentire German nuclear fission project. This oversimpli-fied version of the project's history sat easily with astereotype that was and is widely held by scientists: theview that scientific progress results mainly from a few"great" scientists having profound ideas.

Another commonality between Goudsmit and Heisen-berg, also deeply based in scientific stereotype, had to dowith scientific objectivity and the politicization of science.Goudsmit and Heisenberg tended to classify projectscientists as objective and apolitical, even though theyplainly were doing their work in furtherance of well-defined and well-understood political objectives. Bothmen tended to classify the physicists administering theproject, equally inappropriately, as political hacks.

The wartime projectIn April 1939, a few months after the discovery of fission,the German physicist Wilhelm Hanle delivered a lectureon possible applications of nuclear energy making use of auranium-graphite pile. Georg Joos, a colleague at Gottin-gen, transmitted a report on Hanle's lecture to theMinistry of Education, which in turn passed it on toAbraham Esau, who was in charge of physics at Ger-

PHYSIC5 TODAY JANUARY 1990 5 3Downloaded 21 Sep 2013 to 128.103.149.52. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://www.physicstoday.org/about_us/terms

The German Decision for Heavy WaterThe initial German experiments oncarbon as a moderator in nuclearreactors were done by Paul Harteck inHamburg, who used dry ice, andWalther Bothe in Heidelberg, whoused graphite. Bothe concluded onthe basis of experiments with graphitehe had obtained from Siemens that auranium machine meeting the require-ments of Heisenberg's theoreticalmodel would not work becausegraphite would absorb too many ther-mal neutrons. Subsequent experi-ments by Wilhelm Hanle showed,however, that Bothe's conclusion waserroneous: Hanle showed that eventhe very pure Siemens graphite con-tained boron and cadmium, strongabsorbers of thermal neutrons, andthat these impurities had been lost tothe air when Bothe reduced thegraphite to ash. Hanle reported hisresults to Army Ordnance, which tookthem fully into account in opting forheavy water as a reactor moderator.

Heisenberg already had demon-strated that a uranium machine relyingon graphite would require much moreuranium and much more moderatorthan a machine relying on heavywater. The cost of producing largequantities of graphite that was free ofcadmium and boron seemed prohibi-tively high to Army Ordnance. Sup-plies of heavy water could be muchmore readily assured, and several

Walther Bothe, winner with MaxBorn of the 1954 Nobel Prize inPhysics, was a leading experimentalphysicist on the German fissionresearch project.

German scientists already were highlyexpert in dealing with heavy water. Itwas a German physical chemist, Karl-Friedrich Bonhoeffer, who persuadedNorsk Hydro, a Norwegian company,to begin commercial production ofheavy water after American research-ers discovered in 1932 that heavywater could be separated from light

water by means of electrolysis. KarlWirtz, an assistant to Bonhoeffer, haddone a lot of work with heavy water,and so had Harteck and Klaus Clusius,a physical chemist in Munich whowas a key figure in Germany's uran-ium isotope separation effort. In Au-gust 1940, Robert Dopel demonstrat-ed experimentally in Leipzig thatheavy water would make an excellentmoderator.

By this time Germany had occupiedNorway, and it was soon decided thatthe quickest, most efficient andcheapest means of assuring a largeheavy-water supply would be to boostNorwegian production. Norsk Hydrowas ordered to increase its annualproduction rate from 20 liters to 1metric ton. In 1941 Norsk Hydro wasordered to install a novel catalyticconversion technology designed byHarteck and his assistant Hans Suess.

It was recognized all along thatordinary water might also be a satis-factory moderator but would requirethe use of enriched uranium. But anisotope separation technique devel-oped by Clusius and Gerhard Dickelran into unforeseen problems, andcentrifuge technology could not bebrought to fruition within a meaningfultime. Thus the Germans came tofocus on developing a reactor fueledby natural uranium and moderated byheavy water.

many's Reich Research Council.Meanwhile, Nikolaus Riehl, a former student of Otto

Hahn and Lise Meitner who was working as an industrialphysicist at the Auer Company in Berlin, brought nuclearfission to the attention of the Army Ordnance Office.Independently of Hanle and Riehl, the Hamburg physicalchemists Paul Harteck and Wilhelm Groth wrote to theArmy drawing attention to the possibility of makingnuclear explosives. Their communication was reviewed atArmy Ordnance, probably by Erich Schumann, the head ofresearch, and Kurt Diebner, the resident expert onnuclear physics. Esau and Diebner were PhD physicists,and Schumann was qualified to teach physics at theuniversity level.

Two uranium workshops, in April and September1939 led to the convening of two conferences, in Septemberand October of that year, dedicated among other things tothe feasibility of using nuclear energy and to the theory ofchain reactions. As a result of the conferences it wasdecided by the Army Ordnance Office to distribute workon nuclear fission among several institutes, rather thanconcentrate it at the Kaiser Wilhelm Institute at Berlin-Dahlem. Important assignments were given to theUniversity of Hamburg (Harteck), Leipzig (Robert Dopel,Heisenberg), the Army Research Center at Gottow(Diebner), the Kaiser Wilhelm Institute for MedicalResearch in Heidelberg (Bothe) and the Kaiser Wilhelm

Institute for Physics (Fritz Bopp, Diebner, Carl-Friedrichvon Weizsacker, Karl Wirtz).

In the fall of 1939 the Dutch physicist Peter Debyewas asked to take German citizenship as a condition of hismanaging military research as director of the KaiserWilhelm Institute of Physics. When he declined and leftto take a position at Cornell, Diebner took over theinstitute. In July 1942, after the institute was returned tothe Kaiser Wilhelm Society, Heisenberg was made direc-tor. Esau was the senior official in charge of nuclearphysics at the Reich Research Council from November1942 to December 1943, when he was replaced by WaltherGerlach. Throughout the war Gerlach, Bothe, Heisen-berg, Harteck, Hahn and Klaus Clusius—all scientists ofthe first rank—were involved in the scientific work andthe administration of the German fission project.

The first phase of the project was devoted to establish-ing the feasibility of building a "uranium machine," thatis, a prototype of a reactor capable of producing energy andexplosive material for nuclear weapons. This was accom-plished by the end of 1941, almost exactly in parallel withwork in the United States and Great Britain. At thisjuncture the project was evaluated, just when Germany'sfortunes in the war were changing dramatically as a resultof Germany's failure to subdue Britain, the first Russiancounterattacks and Pearl Harbor.

A 150-page report to Army Ordnance, prepared

5 4 PHYSICS TODAY JANUARY 1990Downloaded 21 Sep 2013 to 128.103.149.52. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://www.physicstoday.org/about_us/terms

Paul Harteck, a physicist at theUniversity of Hamburg, was perhaps themost dynamic and effective member ofthe German nuclear fission project.Exploiting excellent connections withGerman industry, he made importantcontributions to work on moderators,uranium enrichment and reactor design.

probably by Diebner and younger project scientists,strongly recommended an industrial-level effort to build aworking reactor and to produce fissionable materials,despite the admittedly rather distant prospects for actualconstruction of a bomb. In light of Germany's urgentmilitary situation, competing demands on resources, andscarcity of certain critical materials—notably materialssuitable for use as reactor moderators—Army Ordnancedecided against industrialization of the project. This wasthe final verdict, which never was reassessed.

The project was transferred to the Kaiser WilhelmSociety, in keeping with its designation as basic research(albeit kriegswichtig—important to the war), but then wasshifted back again to the Reich Research Council. By nowthe council was part of the sprawling jurisdiction ofReichsmarschall Hermann Goring, the Air Force chief,but Albert Speer, the head of armaments and later warproduction, also took an interest in the fission project.While bureaucratic battles associated with the project'sloose organization occasioned some unhappiness amongproject scientists, the disaffection appeared to haveeverything to do with personal rivalries and woundedvanity and nothing to do with the fundamental objectivesof the project, which were well understood and accepted onall sides.

Goudsmit attacksDuring the course of 1946, Goudsmit published severalpopular articles on physics under Nazism.5 These essayshad both a pedagogical function and a political aim: toillustrate the debilitating effect of fascism on science, andto argue against tight military control of postwar Ameri-can research. Although Goudsmit presented his accountas authoritative, he grossly misrepresented the Germanscientific achievement. Goudsmit's initial distortion ofthe record was unintentional, the result of sloppy re-search, but once he had staked out positions, he refused torecant. Among other erroneous statements, Goudsmitclaimed that the Germans conceived of an atomic bomb asa nuclear reactor gone out of control and that the Germansdid not seriously consider using plutonium for atomic

bombs.Heisenberg and his close friend and younger colleague

von Weizsacker, who often spoke with one voice, had triedto publish an account of the German nuclear fissionresearch as soon as the imprisoned German scientists hadbeen released from confinement in England and returnedto Germany, but they were stopped by the Britishoccupation officials. But the following year Heisenbergreceived permission to proceed with publication of anarticle on the German fission project, which appeared inDie Naturwissenschaften in November 1946.5 Heisenbergcomposed a preliminary draft—which fortunately hassurvived—and sent copies to selected colleagues forcriticism. By this time Heisenberg had seen a copy of the"Smyth report" and thus knew a considerable amountabout the successful American nuclear weapons project.

When Heisenberg's article is compared with sourcesdocumenting the history of the German nuclear fissionproject, several important discrepancies emerge. First ofall, Heisenberg slighted the project scientists and adminis-trators connected to Army Ordnance and those who heldother high positions in the Third Reich's science policybureaucracy—particularly Diebner, Esau and Schumann.For example, Heisenberg gave Gerlach, Esau's successoras head of the German nuclear fission effort, credit forseveral of Esau's innovations and accomplishments.

Scientists like Esau and Schumann were in manyrespects an embarrassment to the rest of their colleaguesin postwar Germany, for they had held highly visiblepositions under National Socialism. Even after the end ofthe war and the revelations about German atrocities,these professionally respectable, if not world-class, physi-cists often admitted their wartime support of NaziGermany unrepentantly. In other words, they representedexactly that from which Heisenberg and others weretrying to distance themselves.

Graphite and plutonium questionsThe second discrepancy that stands out in Heisenberg'sdraft article of 1946, and one that surprised and annoyedBothe, was Heisenberg's assertion that Bothe's measure-

PHYSICS TODAY JANUARY 1990 5 5Downloaded 21 Sep 2013 to 128.103.149.52. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://www.physicstoday.org/about_us/terms

ment of the diffusion length of neutrons in carbon hadbeen a mistake that had hindered the further progress ofthe entire nuclear fission project. Bothe protested, andHeisenberg accepted the rewording that his Heidelbergcolleague suggested. Nevertheless, this theme, "if only wehad tried graphite . . . ," continued to circulate withinHeisenberg's circle and beyond. Eventually Bothe becamea scapegoat—the scientist who had made "the mistake"that had kept the Germans from achieving a chainreaction.

The reasonable and justifiable decision for heavywater and against carbon as a moderator was made oneconomic grounds by those responsible at Army Ordnance,in full knowledge of the potential of carbon as a moderator(see the box on page 54). Why then did Heisenberg claimotherwise? By the time he wrote his draft article, ofcourse, he had studied the Smyth report, and he had notedthat the Americans—in contrast to the Germans—hadused graphite for their nuclear piles. The facts that theAmericans had used graphite and succeeded, whereas theGermans had not and failed, were used to reach thedubious conclusion that the Germans would have succeed-ed, or certainly would have gone much further, if onlyBothe had not made his "error" and they had chosengraphite as their moderator.

A third discrepancy in Heisenberg's printed article—but not in his draft!—concerned von Weizsacker's theoreti-cal discovery of the explosive properties of plutonium. Inthe draft article, Heisenberg mentions von Weizsacker'sdiscovery that an operating nuclear reactor producesuranium-239, whose transuranic daughter products havethe same properties as uranium-235. Heisenberg thuswrote in the draft that an energy-producing nuclear reactorcould be used to produce materials for nuclear explosives.

Apparently one of Heisenberg's colleagues thoughtthat this passage was too explicit, for the published versionwas much more circumspect. Instead of stating flatly thatnuclear explosives can be produced using materials froman operating nuclear reactor, the sentence now allowedthat von Weizsacker's work made it more probable that anenergy-producing nuclear reactor could manufacture fis-sionable materials, but immediately added that the"practical execution" of this process had not been dis-cussed at the time. The form the final version took couldbe taken to imply that this lack of discussion wasintentional.

A fourth important discrepancy was Heisenberg'sdescription of how the Germans came to conclude thatnuclear weapons could not influence the further course ofthe war. Here Heisenberg grossly misrepresented oneimportant aspect of this question by attributing falsely thedecision not to shift the research up to the industrial levelof production to a meeting of a small group of KaiserWilhelm Society scientists with Speer in June 1942.Heisenberg implied that it was Speer who decided that thescientists should focus on the design and construction of a"peaceful" nuclear reactor rather than on militaryapplications.

In fact, the decision to leave the research at thelaboratory scale was made by Army Ordnance alone, morethan half a year before the meeting with Speer. Butapparently Heisenberg preferred to attribute this crucialjudgment to a meeting in which he personally played aconsiderable role rather than to Army Ordnance. Thiswas consistent with his apparent strategy of implying thatproject scientists had actively refrained from working on abomb project for Hitler and of dissociating himself andother project scientists from Army Ordnance, whose

scientists and science policy makers were discredited inthe postwar era.

The alleged conspiracyIn the second-to-last paragraph of the published version ofhis 1946 article, Heisenberg presented an excellentsummary of why the Germans had not attempted theproduction of nuclear weapons on an industrial scale:Until 1942, the Germans anticipated an early end to thewar and were uninterested in weapons that could not beused in the immediate future; after 1942, it was evidentthat a bomb could not be built before the war was over, andGermany's steadily deteriorating military situation ham-pered any further progress.

Had Heisenberg left it at that, his ultimate conclusionabout Germany's fission effort would have been unim-peachable. But both the draft version and the publishedversion ended with a paragraph implying that the Germanscientists—especially the circle around Heisenberg—hadheld themselves back from producing nuclear weapons forthe Nazi state because of moral scruples. Indeed thispassive resistance was supposed to have stopped theGerman production of such weapons. In the publishedarticle, Heisenberg wrote that the German physicists hadstriven from the very beginning to keep the control of theproject in their hands, and that they had used theinfluence they had as experts to steer the research awayfrom the manufacture of nuclear weapons.

Of course it is possible that some of the Germanscientists who worked on nuclear energy and weapons,including Heisenberg, neither intended nor desired thatGerman nuclear weapons be created and used. ButHeisenberg's claim that the scientists willfully hinderedthe creation of nuclear weapons for Hitler's government isnot supported by the documentary evidence, and it isintrinsically implausible. Why should they have fearedand trie.d to prevent that which they knew could not andwould not be done before the end of the war?

Goudsmit's second offensiveGoudsmit was angered after reading a translation ofHeisenberg's article in the British journal Nature. Accord-ing to Goudsmit, Heisenberg's account had all theearmarks of being meant for consumption in Germany: Itwas meant to appeal to national sentiment that Germanscience was good and pure and could not fail.

In Goudsmit's opinion (which was poorly based infact), Heisenberg had not owned up to the failures of theGerman nuclear fission project; indeed he had told a "taleof success." But what really enraged Goudsmit wasHeisenberg's attempt to seduce the casual reader intobelieving that the German scientists had made a "deliber-ate decision" to refrain from making nuclear weapons.Goudsmit argued, correctly, that the Germans themselveshad thought that they were progressing satisfactorily inthat direction.

Heisenberg's article was probably one reason whyGoudsmit decided to take his case to a broader audienceand so wrote the popular book Alsos in 1947. In order thatAlsos not go unnoticed, Goudsmit arranged for it to bepreviewed in Life magazine on 20 October 1947.

Alsos posed and purported to answer a question thatwas of keen interest to Goudsmit and of great topicalimportance at a time when the organization of big sciencewas at issue in the United States: Why did Germanscience fail where the Americans and British succeeded?Goudsmit's answer was that science under fascism wasnot, and probably could never be, the equal of science in a

5 6 PHYSICS TODAY JANUARY 1990Downloaded 21 Sep 2013 to 128.103.149.52. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://www.physicstoday.org/about_us/terms

Uran551 Kg

Paraffin'© Praparat

Alternating horizontal layers of uranium powder andparaffin (diagram, above) were characteristic of the earlyGerman reactor experiments. Heisenberg's insistence oncompleting this kind of experiment late in the war, when

other scientists such as Karl-Heinz Hocker and Kurt Diebnerof Army Ordnance wanted to proceed rapidly with

construction of reactors consisting of uranium latticessuspended in heavy water, may have held back German

progress toward a self-sustaining nuclear reaction. Thespherical design of the early German reactors (right)

apparently fortified Goudsmit in his misconception that thereactors themselves were intended to be bombs. In his book

Alsos he reproduced drawings of such reactors and labeledthem "Germany's atom bomb."

democracy. In Goudsmit's opinion, the "totalitarianclimate" of Nazi Germany led to complacency, politics inscience and hero worship, all of which adversely affectedthe German research.

Goudsmit's account of the Alsos Mission and theGerman nuclear fission research program was basicallythe same as that set out in his earlier articles. Heisenbergwas portrayed as a tragic figure, an extreme nationalistled astray by the Nazis and made to appear foolish by therevelations of Hiroshima. But Goudsmit was concernedwith issues larger than the German nuclear energy andweapons project. Using Gestapo records that he himselfconsidered suspect, Goudsmit unfairly dismissed Schu-mann and other National Socialist science policy adminis-trators as incompetent Nazis and drew an arbitrary lineof demarcation between the "good scientists"—good inboth the professional and moral senses—and the "Naziscientists." Thus Goudsmit did exactly what Heisenberghad done, although his motives were quite different.

Goudsmit's concluding chapter sharply criticizedwhat he saw as American complacency about its scientificand military superiority over the Soviet Union, and heattacked those who wanted to continue wartime restric-tions on nuclear science in the United States. He used theexample of Heisenberg to argue that isolation, secrecy andgovernmental control ruin science.

Heisenberg-Goudsmit correspondenceThe publication of Alsos touched off a fascinating corre-spondence between Goudsmit and Heisenberg. The ex-change brought out a central issue in postwar psycholo-gy—the issue of whether the past should simply be buried,so that one could move on unencumbered, or whether it

needed to be addressed and worked through.Without having read Alsos—indeed there is no

record that he ever read it—Heisenberg wrote to Goud-smit in the fall of 1947 and enclosed a copy of his articlefrom Die Naturwissenschaften. Using polite language,Heisenberg remarked that he had seen several of Goud-smit's articles and had gotten the impression that Goud-smit was unaware both of the details of the Germanfission effort and the psychological situation in Germanyduring the war.

After reminding Goudsmit that he had misled Heisen-berg in the spring of 1945 (when Goudsmit still was subjectto the secrecy requirements of the Manhattan Project) bytelling him that the Americans were not working onnuclear weapons, Heisenberg went on to describe thewartime situation in Germany in a manner difficult toreconcile with his conduct during the war. On one hand,he wrote, it was clear to the scientists what "heinousconsequences" the victory of Nazism in Europe wouldhave, but on the other hand, in view of the hatred towardGermans with which the war had saturated Europe, theycould hardly look forward to the country's utter defeat.The situation, Heisenberg argued, led them to assume amore "passive and humble manner."

After some delay, Goudsmit finally replied to Heisen-berg on 1 December 1947 by criticizing Heisenberg'sjustification of this "passive and humble manner." Goud-smit had been deeply disappointed to learn of Heisenberg'sattempts at a compromise with the Nazis, he said. Whatsurprised him most was that Heisenberg did not see thatsuch a compromise was impossible. Attempts to convincethe Nazis of the soundness of relativity and quantumtheory—an effort Heisenberg emphasized in his opening

PHYSICS TODAY JANUARY 1990 57Downloaded 21 Sep 2013 to 128.103.149.52. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://www.physicstoday.org/about_us/terms

letter—seemed to Goudsmit so out of place.How, Goudsmit asked, could Heisenberg have hoped

to be successful, or have thought that these wereimportant issues? In Goudsmit's opinion, not the periodunder Hitler but the present was the right time for the"more humble manner" Heisenberg described.

Heisenberg replied in early 1948, and this time histone was colder and less polite. First of all, he made itclear that the question he now considered most importantwas whether the Germans had known how an atomicbomb would have worked. He then went on to refuteGoudsmit's claims on this score by showing that theGermans had understood both fast-neutron chain reac-tions and the potential of plutonium. Letting a littlesarcasm leak into his tone, Heisenberg remarked thatobviously Goudsmit had accidentally overlooked thereports that would have given him the correct picture.Only after they had agreed on the "facts" of the Germanscientific achievement would Heisenberg be willing todiscuss the political motives behind the work.

However, Heisenberg wanted to comment on a fewpoints raised by Goudsmit's letter. First of all, he pointedout that he had always believed that German science hadsuffered under Nazism, especially because of the expulsionof many capable scholars from Germany and the advance-ment of nonsensical theories such as deutsche Physik.Moreover, Heisenberg had made critical comments aboutdeutsche Physik in public at a time when such action hadbeen dangerous. Also, it never would have occurred toHeisenberg to think that the German physicists were anydifferent from their Allied counterparts. But how couldGoudsmit continue to overlook the fact that the Germanphysicists also found themselves in a different psychologi-cal situation from their colleagues in England andAmerica?

Heisenberg also commented on what Goudsmit haddescribed as compromises with Nazism by denying that hehad been so naive as to believe that there was much chanceof winning over SS leader Heinrich Himmler, and hebluntly stated that he would have "criminally" neglectedhis duty if he had not, at least in his small circle, tried toshatter the "delusion" of the dictatorship. In particular,Heisenberg had never had the slightest sympathy for thepeople who withdrew from all responsibility during theThird Reich but then in a safe dinner conversation wouldtell someone that National Socialism would ruin Germanyand Europe, just wait and see.

At the end of his letter, Heisenberg turned to thepresent situation in Germany and his views on dealingwith the past. It was difficult, he wrote, to win the heartsand minds of people through the force of arms, especiallybecause of the "indescribable misery" in Germany. Whatthe Germans needed was not a hateful settling of accountswith the past, but instead a quiet reconstruction of a lifeworthy of a human being. In any case, Heisenberg assuredGoudsmit that he could be certain that the Germanphysicists would gladly participate in any effort thatwould contribute to a "better world understanding."

Intervention by van der WoerdenAt this point the Dutch mathematician Bartel L. van derWaerden, a friend of Heisenberg, former countryman ofGoudsmit and a very perceptive observer, briefly enteredthe debate. Van der Waerden had spent the war years atthe University of Leipzig, where for many years he wasHeisenberg's colleague. After the war he went to teach atJohns Hopkins, where he read A Isos in March 1948 and ex-perienced the American debate over the book firsthand.

Subsequently he wrote both Goudsmit and Heisen-berg. Van der Waerden told Goudsmit that there was one

point in Alsos that he did not understand. Did Goudsmitmean, he asked, that the German physicists, knowing whoHitler was, had planned "the horrible crime" of putting anatomic bomb into his hands? Goudsmit replied (obscurely)that such an act would not have been a crime; rather itwould have been something that could not be stopped.(Perhaps he had in mind the "inevitability" that the USweapons scientists often talked about, once the feasibilityof an atomic bomb was established in the ManhattanProject.)

According to Goudsmit, the Germans thought thatmaking an atomic bomb was much more difficult than itactually was, and thus the question of conscience was notso urgent for them. This argument pervaded all ofGoudsmit's publications on the German nuclear fissionproject and was an attempt to reconcile the evidence thathe had gathered with his profoundly ahistorical andnoncontextual preconceptions of how science and technol-ogy work. In his view, since the Americans had succeededin building an atomic bomb, the Germans should somehowhave known that it was feasible. Since the Germanscientists and authorities had decided that this task wasnot feasible, Goudsmit concluded that this decision musthave been a mistake, and searched until he thought he hadfound the German error in the concept of a nuclear pile asa bomb—an error they never made.

(In November 1944 the Alsos Mission heard from asecondhand source that Hahn's institute had estimatedthe minimum mass of an atomic bomb at eight tons. Inhindsight it is evident that the source was confusing themass of a reactor with the mass of a bomb, but Goudsmitattributed the confusion to the whole project; a subsequentmisreading of a letter by Gerlach fortified Goudsmit in hisconviction that the Germans thought the reactor itselfwould be the bomb.)

Van der Waerden wrote Heisenberg immediatelyafter his conversation with Goudsmit, noting that docu-ments in Goudsmit's office verified Heisenberg's claimsabout what the Germans had known about plutonium andnuclear weapons, but adding that in his opinion questionssuch as the "complacency" of the German physicists orwhat Heisenberg and other scientists had understood oroverlooked were insignificant.

The two Dutch-born scientists also had discussed the"question of guilt," van der Waerden told Heisenberg. Inthe end they still disagreed over the psychologicalquestion—what Heisenberg's group would have done ifthey had made greater progress—but agreed someoneshould not be condemned for what he might have done ifthe situation had been different.

Although van der Waerden was defending Heisen-berg, he was also critical of his friend and formercolleague. In a second letter, sent the very next day toGermany, van der Waerden asked for more information.Heisenberg had informed high-ranking German authori-ties about the potential of nuclear explosives during thewar; had he also considered the question of responsibilityat that time? Had Heisenberg's statements touting themilitary potential of the project been a mere ruse to getmoney for physics?

As Heisenberg's "lawyer" (van der Waerden's term),van der Waerden said he had enough evidence to defendhim, but as Heisenberg's friend he also wanted to believethat under any circumstances Heisenberg's decency wouldhave been stronger than the combination of his national-ism and ambition.

Van der Waerden made one last attempt to reconcileGoudsmit and Heisenberg before he returned to Holland.Naturally Heisenberg was right to claim, van der Waer-den argued in his second letter, that his efforts on behalf of

5 8 PHYSICS TODAY JANUARY 1990Downloaded 21 Sep 2013 to 128.103.149.52. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://www.physicstoday.org/about_us/terms

Von Laue ond Morrison on the Arming of 'Himmler ond Auschwitz'

Coudsmit's book Alsos touched off an emotional debate inthe Bulletin of the Atomic Scientists. Philip Morrison, aManhattan Project scientist who had also been involvedwith scientific intelligence, reviewed the book and attackedHeisenberg's suggestion that the Germans had not wanted tocreate nuclear weapons. Like their counterparts in Americaand Britain, Morrison stated, the German scientists hadworked for the military as best as their circumstancesallowed. The difference, which Morrison found unforgiva-ble, was that they worked for the "cause of Himmler andAuschwitz," for the burners of books and the takers ofhostages.

With justification, Morrison and others in America be-lieved that the Germans were implying that German scien-tists had been morally superior to their American and emigrecounterparts because they had not built a bomb—andbecause in particular, they had not built bombs that wereactually used to destroy cities.

Morrison's attack provoked a response from Max vonLaue, perhaps the only German physicist residing in Ger-many who still commanded respect in America at that time.Von Laue was a more outspoken critic of National Socialismthan Heisenberg or von Weizsacker, but as acting director ofthe Kaiser Wilhelm Institute for Physics, he had helpedoversee the work on applied nuclear fission and was awareof its military potential. Von Laue's reply to Morrison waspublished in German in Physikalische Blatter, so that manyGerman readers saw only his forceful comments and not theoriginal critiques by Goudsmit and Morrison.

Von Laue attacked Morrison for the "monstrous sugges-tion" that German scientists as a body had worked forHimmler and for Auschwitz. He recognized that Goudsmithad lost many of his closest relatives in concentration campsand what unutterable pain the mere word "Auschwitz" mustevoke in the Jewish physicist. But von Laue then went on todraw the conclusion that neither Goudsmit nor his reviewerMorrison was capable of an unbiased judgment.

Just because a few German scientists had managed toavoid being drawn into the maelstrom, von Laue wrote, thisdid not mean that all could have. Von Laue argued thatwhereas open refusal to participate in war work would haveled to catastrophic personal consequences, a fictitiouscompliance often allowed German physicists to shieldyounger researchers from the war. Sometimes, von Lauecontinued, the physicists protected "political suspects" fromconcentration camps by assigning them work labeled "ofmilitary importance." Von Laue asked ironically whethersuch scientists and internal refugees should be labeled"armorers of Himmler and Auschwitz." Von Laue said thatarticles such as Morrison's merely kept "alive hate."

Atomic Scientists

Regretting that he had to take issue with von Laue's"moving statement," Morrison pointed out in the same issuethat he had accused the German scientists of working not forHimmler, but for Himmler's cause, the victory of NaziGermany.

For Morrison, Goudsmit and many people who had beenin or had fled from countries dominated or threatened byGermany, working for a German victory in World War II wasseen—after the fact—as equivalent to working for the causeof Himmler and Auschwitz. But, during the Third Reich fewGermans had distinguished clearly between the knownmilitary and political goals of the National Socialist govern-ment and those of Germany or the German people. Oncethe full extent of Nazi crimes was revealed and undeniable,Germans such as von Laue and Heisenberg advanced aretrospective argument implying that efforts toward a Ger-man victory and for the good of the German people hadbeen separable from allegiance or service rendered toNazism.

German physics represented a significant success. But onthe other hand, van der Waerden could understand thenegative reaction of Goudsmit and others toward thosesame efforts. This negative reaction was admittedlyillogical. But emotionally it was comprehensible. CouldHeisenberg associate with the SS and leading NationalSocialists, even exert influence over them, van derWaerden asked, without compromising himself? Van derWaerden believed that he could, but could understand ifothers did not feel that way.

In a letter written on 28 April 1948, Heisenbergreplied to van der Waerden by addressing the moral

question once again. When he knew near the end of 1941that a nuclear reactor would work and that nuclearweapons probably could be built, Heisenberg explained, hehad been "shocked" by the thought of such weapons in thehands of some ruler, and not only Hitler. In any case,Heisenberg flatly stated, he would have considered it acrime to make atomic bombs for Hitler. But Heisenbergalso considered it unfortunate that these weapons weregiven to other rulers and were used by them. During thepast few years, Heisenberg said, he had learned somethingthat his friends in the West did not want to understand:During times like those in Germany during the war,

PHYSICS TODAY JANUARY 1990 5 9Downloaded 21 Sep 2013 to 128.103.149.52. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://www.physicstoday.org/about_us/terms

hardly anyone can avoid committing crimes or supportingthem by doing nothing. Heisenberg hastened to add thathe did not mean he would have been prepared to commitany sort of crime for Hitler.

A failure to communicateGoudsmit wrote Heisenberg in late September of 1948 andadmitted that he had been wrong about certain details, butnow he grasped mistakenly at another perceived Germandeficiency by claiming that the Germans had not under-stood that an atomic bomb would rely on a fast-neutronchain reaction. However, Goudsmit's main aim in thisletter was again to stress how science had suffered underNazism and how political interference ruins science. AsGoudsmit himself realized, the contents of this letterdiffered little from the previous one he had sent Heisen-berg.

After spending several letters discussing the moralaspects of the German nuclear fission project, only to haveGoudsmit stubbornly continue to make false statementsabout Germany's scientific achievements, Heisenbergdealt almost exclusively with the question of scientificcompetence in replying to Goudsmit's latest missive. TheGerman physicist noted with pleasure Goudsmit's grudg-ing partial admission of error. Now that Goudsmit hadagreed that the Germans had known about plutonium,Heisenberg wanted him to make this admission public, inthe Bulletin of the Atomic Scientists.

But Heisenberg did more than gloat over Goudsmit'sdiscomfiture. Naturally, he wrote, he agreed that atotalitarian system greatly damages science, but in Alsosthis conclusion was based on false arguments, whichHeisenberg found very unfortunate.

Goudsmit found Heisenberg's answer "impertinent"in turn and could not understand why the recognition ofthe value of Heisenberg's scientific work meant so much tohis German colleague. Heisenberg may well have beenasking himself why Goudsmit could not recognize theadmittedly modest German achievement.

Before Goudsmit could write his German colleague orpublish anything in the Bulletin, Heisenberg took his casedirectly to the American people by means of an interviewpublished in The New York Times. He devoted most of theinterview to refuting Goudsmit's claims about the Germanscientific work, but he also addressed the moral questioncarefully: Because of their sense of decency, most of theleading German scientists had disliked the totalitariansystem, he claimed; yet as patriots, when called upon towork for the government, they could not refuse.

Goudsmit responded in a letter to the Times byadmitting that his portrayal in Alsos was an oversimplifi-cation, but he still insisted stubbornly that the Germanshad had only a very vague notion of how an atomic bombworks.

On 11 February 1949 Goudsmit wrote Heisenbergagain, advancing the same arguments, and in an exasper-ated tone asked whether further correspondence madeany sense. Heisenberg responded with his most curt letteryet, stating bluntly that he would have preferred no publicdiscussion of the German nuclear energy and weaponsresearch, that through his articles and book Goudsmitrepeatedly had spread false information about the Ger-man work, that it was time that a correct description of theGerman research was presented in the newspapers for achange, and that he was disappointed that Goudsmit didnot recognize Heisenberg's right to take his case to thepublic as Goudsmit had done so often. It is clear that inthe end Goudsmit's unfair criticism of Heisenberg'sscientific abilities and achievement exasperated andembittered the German physicist. The damage done to his

reputation as a physicist may have come to bother himmore than the criticism he received for serving theNazism. The debate between Goudsmit and Heisenbergleft outsiders with the impression that Heisenberg wasmorally insensitive or even obtuse, but that impression isnot altogether warranted. Private wartime correspon-dence shows that Heisenberg was indeed troubled by hiscompromises with National Socialism. He also wastroubled by the fate of individuals. In 1943, at the requestof a Dutch physicist, Heisenberg wrote a letter toauthorities on behalf of Goudsmit's parents, who wereabout to be sent to a concentration camp. Even for aperson in Heisenberg's position, this was a very dangerousmove at that time. Yet throughout the long and painfulcorrespondence with Goudsmit after the war, Heisenbergnever mentioned his intervention on behalf of Goudsmit'sparents.

Last letters and visitIn the course of their protracted debate, both Goudsmitand Heisenberg had come very close to losing theirtempers, and they now realized that it was pointless tocontinue. Goudsmit wrote one last letter in which herepeated his litany of arguments, told Heisenberg that hewould not bring up the matter again, and expressed thehope that they could continue to correspond about physics.Heisenberg replied immediately and thanked him for hisletter, which this time Heisenberg had been very happy toreceive. Goudsmit could be certain, Heisenberg promised,that he would not stir up the controversies of the past fewyears again, and especially not in public. Exactly likeGoudsmit, Heisenberg said, he believed that it was moreinteresting to discuss physics than the unpleasant past.

Both men worked very hard the rest of their lives torestore international cooperation in science. Despite theloss of his parents, Goudsmit went out of his way to assistGerman scientists. In the same issue of the Bulletin of theAtomic Scientists that contained a debate between PhilipMorrison and Max von Laue on the German scientists' roleduring the war (see the box on page 59), Goudsmitcontributed an essay that showed he had changed his mindabout the sink-or-swim policy he once had advocated forGerman science. Entitled "Our Task in Germany,"Goudsmit's article once again called for internationalscientific cooperation, and this time he extended his handto Germans as well. Goudsmit expressed the belief thatAmerican scientists should morally support those Germancolleagues who were worthy of confidence, and he addedthat there were many of them. Americans did not have toagree with all of their opinions but should makeallowances for the disturbing circumstances under whichthese Germans had lived and were still living. Americanshad to communicate with them, Goudsmit maintained, asin the days before Hitler. It was as if Goudsmit had beenconverted by the disturbing forces that he had helped tounleash.

In 1950 Heisenberg visited Goudsmit at BrookhavenNational Laboratory in Long Island, New York, and theydiscussed only physics.

References1. M. Walker, Vierteljahrshefte fur Zeitgeschichte 38, 1 (1990).2. A. Beyerchen, Scientists under Hitler, Yale U. P., New Haven,

Conn. (1977). R. Rhodes, The Making of the Atomic Bomb,(Simon and Schuster, New York (1988).

3. J. Radkau, Aufstieg und Krise derdeutschen Atomwirtschaft,1945-75, Rowohlt, Reinbeck (1983).

4. S. Goudsmit, Rev. Sci. Instrum. 17, 49 (1946); Bull. At. Sci 1, 4(1946); Sci. Illustrated 1, 97 (1946).

5. W. Heisenberg, Naturwissenschaften 33, 325 (19461. •

6 0 PHYSICS TODAY JANUARY 1990Downloaded 21 Sep 2013 to 128.103.149.52. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://www.physicstoday.org/about_us/terms