Sometimes the Records Tell Different Stories

Today’s post is written by Dr. Greg Bradsher, Senior Archivist at the National Archives at College Park

History is the version of past events that people have decided to agree upon.

 Napoleon Bonaparte

The History of our Revolution will be one continued Lye [lie] from one end to the other. The essence of the whole will be that Dr. Franklins electrical Rod, smote the Earth and out sprung General Washington. That Franklin electrified him with his rod—and thence forward these two conducted all the Policy, Negotiations, Legislatures and War.

Letter, John Adams to Benjamin Rush, April 4, 1790

The past is the past. History is what someone says about what happened in the past. Historians, and others, consult textual records, oral histories, non-textual records, and artifacts to find evidence of the past. Needless to say, persons writing about people, places, and things observe and/or record those things from their own perception and sources at hand, which might be their own eyes and ears. Thus, it is understandable that two people witnessing the same thing might have a different view of what they saw or heard. To some degree, this should be just common sense to everybody, but it is useful to be periodically reminded of this.

Just look at the stories regarding Judas Iscariot in the New Testament. Almost no two accounts agree on his motivations, his actions, the circumstances of him receiving thirty pieces of silver to betray Jesus, and the circumstances of his death. Of course, the authors of the Gospels were writing many decades after the events they recorded, and we do not know what their written and oral sources were.[1] I found in doing research on the death of Adolf Hitler that rarely do those in the Berlin bunker with Hitler during his last days, consistently record or relate what happened. And what they related was often weeks after the events. As a result, I wrote what I thought most likely happened and when it happened.[2] I did so keeping in mind what was written almost sixty years ago: “The historian can rest satisfied even when his explanation is not absolutely probable, provided he has shown that it is significantly more likely than any of the comparable alternatives.”[3]

When faced with conflicting accounts regarding the discovery of Nazi gold reserves in the Merkers Mine in Germany in April 1945, I ended up writing in my article about the subject:

Early the next morning [April 6], two military policemen guarding the road entering Keiselbach from Merkers saw two women approaching and promptly challenged and stopped them. Upon questioning, the women stated that they were French displaced persons. One of the women was pregnant and said she was being accompanied by the other to see a midwife in Keiselbach.[4]

These women related information about a mine in Merkers holding treasures. In researching for the Merkers Mine article, I found three reports about the above incident, all written within 48 hours of the events they recorded. There were inconsistencies regarding the nationalities of the women, which direction they were headed, and whether one of them was a midwife. I selected to use the one that seemed the most likely.

Again, in writing about the capture of one of Hitler’s couriers who was able to escape Berlin with copies of Hitler’s personal will, political testament, and his marriage license, I ran into vague and conflicting information about how the individual, SS Col. Wilhelm Zander, was captured. In my Prologue article I gave a barebones account:

The Americans captured Zander and his documents (including the original marriage license of Hitler and Braun, and the handwritten transmittal letter from Bormann to Doenitz) with the assistance of British intelligence officer Maj. Hugh Trevor Roper, in Bavaria on December 28. [5]

In a blog, dealing with the Hitler documents and the capture of Zander, I was more detailed. I wrote:

With the lead to Aidenbach, Trevor-Roper, accompanied by Weiss, and apparently a CIC officer named Rosener, in a jeep set out from Munich on the night of December 27 for the 90-minute drive to Aidenbach. Clearing through the Regional CIC office and the Degendorf Sub-Regional Office, where an agent named Brickmann joined them, sometime between 3am and 4am on December 28 they found the farmhouse where Zander was supposedly staying. Trevor-Roper posted an American soldier with a revolver at each corner, and knocked on the door. There was no answer. Trevor-Roper ordered a German policeman to climb through the window and open the door. Inside, they found a man in bed who claimed to be a merchant named Wilhelm Paustin. With him was Ilsa Unterholzner. Both were arrested. Trevor-Roper made them dress, and then, with Weiss, drove them back to Munich for interrogation.[6]

If one looks at my footnote citation to the above information it will be seen that I relied on some dozen sources to tell the story. Each source provided somewhat different versions of events. For example, CIC agent Arnold H. Weiss, who was with Trevor-Roper later recalled that as the Military Police broke down the door, a shot rang out from the house. The Military Police found the startled Zander naked in bed with a woman and quickly overpowered him. Weiss grabbed Zander’s Italian Beretta-a memento he kept. Weiss told Zander they had come to arrest him and asked him his name. He said Paustin produced an identity card. Weiss said it was a fake and he was taken into custody and taken to Munich. [7] This account, recalled sixty years after the event, is more dramatic than the other accounts, which were written shortly after the events occurred. Some or all of Weiss’ account may be true, but I decided to tell the story based on a blending of other evidence, which was less dramatic.

In 2016, a radio producer and I spent considerable time researching the records at the National Archives regarding the May 1945 capture of the infamous Nazi, Julius Streicher. We were trying to determine whether Streicher was captured by Major Henry Plitt and his colleagues or by Werner Moritz. The records provided relatively consistent information about the date and the place of capture, but there was actually almost no documentation regarding who actually captured Streicher and what transpired in the immediate aftermath. The producer on her radio program ended up saying “The specific records about Streicher’s capture are spotty and contradictory. Usually someone else is credited with capturing him. But we know Werner was in the area doing this kind of work at the time. And an expert at the National Archives told me he does think Werner could have done this.” [8] The latter is a reference to Moritz capturing and mistreating Streicher.

Sometimes people provide intentionally or unintentionally untruthful and misleading information about themselves. Take the case of Iver Olsen. When I first came across his name in the records, he was identified as being a special representative of the War Refugee Board [WRB] at the American Legation in Stockholm and in contact with Raoul Wallenberg. Subsequently I learned that he had connections with the United States Treasury Department and the Office of Strategic Services [OSS]. A joint Swedish-Russian study of Olsen’s ties to Wallenberg said “It is also clear that the OSS and WRB cooperated closely with each other and that Iver Olsen represented the OSS first and foremost. His other assignments were mainly to give him ‘cover.’ According to the CIA, Olsen sometimes mixed his OSS duties with his WRB activities.” [9] The significance of Olsen’s official and cover relationship with Wallenberg was what the Soviets believed. According to the study:

A CIA paper from 1990 summarised the organisation’s knowledge of Raoul Wallenberg’s indirect connections with the OSS. This was the first time that Iver Olsen was officially acknowledged as also working for the OSS. This document drew the same conclusion about Raoul Wallenberg as that mentioned above. Thereafter followed a proposal for two alternative courses of action for the US government: 1) to say nothing beyond what had already been said in previous official statements. To say more would embarrass Wallenberg’s supporters and family by suggesting that Raoul Wallenberg may have been undertaking intelligence-related assignments despite Olsen’s statement in 1955 to the contrary; 2) to initiate unofficial contact with the Soviet Union and be more forthcoming about Iver Olsen. The main advantage of this would perhaps be to discover how the Russians came to suspect Raoul Wallenberg. The author doubted whether much would be gained from the latter alternative. A quotation from the same paper stated unequivocally, ‘Strange as it may seem to the present generation of intelligence officers, the document (from 1955) confirms Olsen’s belief that he was working for three government agencies simultaneously … that his contacts with Raoul Wallenberg had nothing to do with undercover activities but took place under the supervision of the WRB’. [10]

Gerald Mayer is another person who had a cover during World War II. In an article about Fritz Kolbe, who worked in the German Foreign Office during World War II and supplied intelligence to the Office of Strategic Services in Bern, Switzerland, I wrote that Mayer was working for the Office of War Information:

Dulles rented a flat at 23 Herrengasse in the picturesque medieval section of Bern, near its cathedral and casino, and placed an inconspicuous sign outside his door: “Allen W. Dulles, Special Assistant to the American Minister.” The flat had a back entrance that people could use when they came to see him at night. For those who might come to Dulles’s front door in the evening, he pulled some strings and had the streetlight opposite his front door turned off for the duration of the war. That Dulles was able to pull the strings probably related to the fact that he had only been in Bern for a few weeks when one of its most respected and widely read newspapers published an article describing him as ‘the personal representative of President Roosevelt’ with a ‘special duty’ assignment. Most of Dulles’s surreptitious important visitors were duly noted by the Swiss, and this resulted in a ‘great man’ image that filtered down through the bureaucracy. Swiss bureaucrats treated him with some deference.

It was no real secret that Dulles was an intelligence operative or that his home was being used for intelligence work. However, a circumspect concern for Swiss sensibilities dictated that he at least seek an office that could claim diplomatic immunity, so he set up shop at 24 Duforstrasse with Gerald Mayer, whose OWI propaganda operation had taken office space on the first floor of the building. Dulles occupied the second floor, with a very small staff.

Because of the related nature of their work and because Dulles was badly understaffed, he very quickly ‘recruited’ Mayer to assist him with OSS work. Mayer was the chief of the OWI Switzerland outpost. He oversaw OWI operations in Switzerland, which consisted of the disseminating OWI and United Nations material in that country and surrounding occupied territories, gathering intelligence material, and analyzing United States propaganda efforts. Mayer, of German-Jewish extraction, had spent much of his life in Europe and spoke fluent German. [11]

Later, I learned that Mayer, who was given OSS code number 678, had been sent to Bern in the spring of 1942 by the OSS, with a cover of him working for the Legation. William J. Donovan, the OSS head, wrote the State Department in March 1942, that the Legation should pay his salary and expenses, and that OSS would reimburse the State Department.[12]

Also, in writing an article about Fritz Kolbe, I ran across several times references to Dr. Ernest (also Ernesto) Kocherthaler. I felt confident in writing:

Kolbe had met Kocherthaler in the mid-1930s while serving in the German embassy in Madrid and sought his assistance in making contact with the British. Kocherthaler, a German-born Jew and businessman, left Germany in the middle 1920s and went to Madrid, where he took over some of his father’s business interests there and also engaged in importing Russian oil into Spain. [13]

In doing the research for the article I found in the records of the American Consulate in Bern, Kocherthaler’s 1947 application for a nonimmigrant visa. I laughed when I reviewed the application because he had listed his birthplace as Spain, not Germany. He probably thought he would have fewer, or no, questions asked of him if he indicated that he was born in Spain rather than in Germany. Fortunately, by the time I saw the application, I had seen enough evidence in the records to indicate that he was indeed born in Germany.

Another instance in the National Archives’ holdings of a person providing different and misleading information about themselves, involves the famous dancer Isadora Duncan. The National Archives holds at least fourteen passport applications submitted by Duncan. In the first one, issued October 10, 1904, she wrote that she was born on May 27, 1879. On subsequent applications she indicates that she was born May 27, 1880. On the first application she stated that she left the United States in June 1900. On the second one, issued December 15, 1906, she indicated that she left the United States in April 1900. On her third application, issued November 2, 1907, she indicated she left the United States March 1889 [she probably meant 1899]. Her first application indicated her height was 5’6”. On the second one she indicated that her height was 5’5”. On her third application she indicated she was 5’4” tall. On her fifth application, issued January 4, 1913, she says she is 5’4”. On her seventh and ninth applications, issued in 1916, she indicates she is 5’ 5 ¾”. In her tenth application, issued February 5, 1918, she says she is 5’6”. On the different applications she lists her occupation variously as “Artist,” “Dancer,” “Artistic Dancer,” and “Theatrical Artist.” It is interesting to note that Wikipedia indicates she was born May 26, 1877 or May 27, 1878, and notes in a footnote “While her birth date is widely given as May 27, 1878, her posthumously-discovered baptismal certificate records May 26, 1877.”

Numbers recorded in the records, and the interpretation of them, can also present a challenge. In 1989, Canadian writer James Bacque, using National Archives sources, published a book, Other Losses, in which he alleges that General Dwight D. Eisenhower intentionally caused the deaths by starvation or exposure of around a million German prisoners of war held in Western internment camps briefly after World War II. For an interesting discussion of the accuracy of Bacque’s numbers see Wikipedia: Other Losses.

In reviewing some text that we planned on adding to NARA’s website in conjunction with albums containing photographs depicting looted art work, a colleague, Robin Waldman, had a comment after she looked at the following:

[Colonel Robert Storey, an American prosecutor at the International Military Tribunal at Nuremberg, addressing the court on December 18, 1945, after having introduced 39 albums as a United States Exhibit]

I should like to refer, while Your Honors are looking at these [the albums], just to the aggregate totals of the different paintings. Here are the totals as shown by Document 1015(b)-PS, which is in the document book. As they are totaled, I don’t think Your Honors need to follow the document; you can continue looking at the books [albums] if you like.

[citing an Einsatzstab Reichsleiter Rosenberg (ERR) report]

Up to 15 July 1944 the following had been scientifically inventoried:

21,903 Works of Art: 5,281 paintings, pastels, water colors, drawings; 684 miniatures, glass and enamel paintings, illuminated books and manuscripts; 583 sculptures, terra cottas, medallions, and plaques; 2,477 articles of furniture of art historical value; 583 textiles (tapestries, rugs, embroideries, Coptic textiles); 5,825 objects of decorative art (porcelains, bronzes, faience, majolica, ceramics, jewelry, coins, art objects with precious stones); 1,286 East Asiatic art works (bronzes, sculpture, porcelains, paintings, folding screens, weapons); 259 art works of antiquity (sculptures, bronzes, vases, jewelry, bowls, engraved gems, terra cottas).

Storey told the judges that they did not have to look at United States Prosecution Document 1015(b)-PS as the numbers were totaled. Robin did look closely at the numbers, and contacted me with the comment, “Greg, The numbers don’t add up.”  She was correct.

So, I looked at the published transcript of the tribunal (Volume 4, page 89) and found it contained the same apparently incorrect information. Then I went to the stacks and looked at the original unpublished transcript. It was the same text.

I then asked my colleague Dr. Sylvia Naylor if she wanted to solve, if possible, the inconsistencies between the total number of art works and the sub-totals. She agreed. Before we went to the stacks to look at United States prosecution document 1015(b)-PS, I took a look at Volume 1 of Nazi Conspiracy and Aggression, a multi-volume work produced after the Nuremberg trial in 1946 by the staff of the Office of Chief of Counsel for the Prosecution of Axis Criminality. I believed the chapter entitled “The Plunder of Art Treasures,” might have some explanation why the numbers did not add up. Here is what was written:

(4) Works of Art (West). The Robert Scholz report declared that:

‘During the period from March 1941 to July 1944, the Special Staff for Pictorial Art brought into the Reich:

29 large shipments including 137 freight cars with 4,174 cases of art works.’ (1015-PS)

The report stated that a total of 21,903 art objects of all types had been counted and inventoried, and stated:

With this scientific inventory of a material unique in its scope and importance and of a value hitherto unknown to art research, the Special Staff for Pictorial Art has conducted a work important to the entire field of art. This inventory work will form the basis of an all-inclusive scientific catalog in which should be recorded history, scope and scientific and political significance of this historically unique art seizure. (1015-B-PS)

The following is a summary of the inventory attached to the report:

Hand-made art objects………5,825
East Asiatic objects……….1,286

The report stated that the above figures would be increased since seizures in the West were not yet completed and it had not been possible to make a scientific inventory of part of the seized objects because of the lack of exports. (1015-B-PS)

The total was still 21,903, but the number of paintings had increased from 5,281 to 10,890. So Sylvia and I went to the stacks to look at File 1015(b)-PS. We reviewed both a translation and a photostat of the original German-language report by Scholz. They both indicated 5,281 paintings. Now I know one cannot always trust numbers in government reports, of any country, but if the total was indeed 21,903, the number 10,890 found in the published volume cited above seemed to be the correct number to make the sub-category totals add up. As long as we were in the stacks we decided to look at the French language version of File 1015(b)-PS, which the French prosecutors had submitted as a French exhibit (RF-1323). We found that it was the same, except in French, as the document in the United States prosecution files. So we went back to the folder containing File 1015(b)-PS. We found File 1015(GG)-PS, which was identified as an undated and unsigned inventory of artworks. The document did not indicate it was an attachment to some other document. In translation it contained the same sub-category numbers as in the File 1015(b)-PS, but provided the following information, regarding paintings:

Oil Paintings 3,027
Water-colors 766
Drawings 1,332
Engravings 4,525
Miniatures-portraits 442
Miniatures-parchments 51
Persian miniatures 49
Glass paintings 19
Porcelain-enamels-paintings 455
Books and manuscripts 55
Reproductions 13

Total 10,890

So part of the mystery was solved regarding the total number of paintings, a number with the others that would add up to 21,903. But we could not figure out what combination of numbers above were used to make 5,281 in the actual report. The only clue was a footnote to Engravings that indicated: “In enumerating engravings in portfolios, the sheets were counted individually. In bound volumes the engravings were consider as one number.” I guess it is possible, if one knew the number of portfolios and number of bound volumes, one could make some combination of numbers to come up with 5,281.

Just to make sure I had not missed anything I went back to Volume 1 of Nazi Conspiracy and Aggression, to the chapter entitled “The Plunder of Art Treasures.” I looked at the end of the chapter where relevant documents are listed, along with the volume and page where they are reprinted in translation. It provided the following information:

1015-GG-PS Inventory of art objects-attached to a report (Document 1015-B-PS), Volume III, p. 670.

I quickly looked at that volume and page number and found the translated document. It was the same as 1015-GG-PS in the stacks in the folder labeled 1015-PS. I then went back to Volume I of Nazi Conspiracy and Aggression and looked at what was said about the document. There in black and white was “The following is a summary of the inventory attached to the report.” I probably should have looked at it more closely the first time I looked at it, but I was more focused on the number 10,890. So, File 1015-GG-PS was an attachment to File 1015-B-PS. That being the case, the numbers do add up. It is unfortunate that Colonel Storey before the International Military Tribunal in December 1945 did not reference File 1015-GG-PS, but perhaps in 1945 it was not clear that the inventory was an attachment to the report. However, in 1946, when Volume 1 of Nazi Conspiracy and Aggression was published the link between the two documents was made.

So what was learned from this exercise regarding the numbers? First, one should not necessarily trust any government-produced numbers, published or otherwise. Second, there is no substitute for looking at the records, even if to verify published versions of it. And third, it pays to look carefully at what you are reading.

Needless to say, there are tens of thousands of documents in our holdings that record the same event differently and/or intentionally or unintentionally misleading. Being aware of that is crucial to understanding the past and being able to write accurately about what indeed happened, knowing that it is almost impossible to be one hundred percent certain that the story you think you know is indeed the true or most factual one.


[1] For an interesting analysis see Tom Bissell, Apostle: Travels Among the Tombs of the Twelve (New York: Pantheon Books, 2015), pp. 3-38. Also useful is

[2] The first of my 8-part blog, entitled “Hunting Hitler Part I-The Bunker (April 28-29).

[3] Nicholas Rescher and Carey B. Joynt, “Evidence in History and in the Law,” The Journal of Philosophy, Vol. 56 No. 13 (June 18, 1959), p. 564.

[4] “Nazi Gold: The Merkers Mine Treasure,” Prologue: Quarterly of the National Archives and Records Administration Vol. 31 No. 1 (Spring 1999)

[5] “Hitler’s Final Words: His Political Testament, Personal Will, and Marriage Certificate: From the Bunker in Berlin to the National Archives,” Prologue: Quarterly of the National Archives and Records Administration Vol. 47 No. 1 (Spring 2015). 

[6] The Search for Hitler’s Political Testament, Personal Will, and Marriage Certificate, Part II

[7] Matthew Brzezinski, “Giving Hitler Hell,” The Washington Post Sunday Magazine, July 24, 2005.


[9] “Raoul Wallenberg: Report of the Swedish-Russian Working Group,” Department for Central and Eastern Europe, Ministry for Foreign Affairs, Stockholm, Sweden, 2000, p. 44.

[10] “Raoul Wallenberg: Report of the Swedish-Russian Working Group,” Department for Central and Eastern Europe, Ministry for Foreign Affairs, Stockholm, Sweden, 2000, p. 41.

[11] “A Time to Act: The Beginning of the Fritz Kolbe Story, 1900 – 1943,” Prologue: Quarterly of the National Archives and Records Administration Vol. 34 No. 3 (Spring 2002).

[12] Donovan letter to the State Department, March 1942, Folder 76, Box 23, COI/ OSS Central Files, Entry 92, Records of the Office of Strategic Services, Record Group 226.

[13]  “A Time to Act: The Beginning of the Fritz Kolbe Story, 1900 – 1943,” Prologue: Quarterly of the National Archives and Records Administration Vol. 34 No. 1 (Spring 2002).


5 thoughts on “Sometimes the Records Tell Different Stories

  1. Today is the 50th anniversary of the Pueblo Incident. In 1968, AP wrote a story that said the U.S.-Soviet conflict off the coast of the North Korea was a “cover-up.” They didn’t know the half of it. So I contacted the AP yesterday, explaining why the Pueblo Incident was not a small cover-up but rather a giant cover-up, and that I was writing a book on the subject with a Russian journalist. No response yet and maybe there won’t be one.

  2. Thank you, Greg, for extrapolating on a point that folks conducting historical research always need to bear in mind: sources are sometimes fragmentary, inconsistent, or contradictory; and one should be cautious about taking any one piece of evidence at face value. Triangulating these sources can be quite challenging, as shown by your examples. Errors made as a result of not taking these factors into account, as any thorough historian knows, have often crept into the literature, to be repeated again and again. To take but one example, most biographies of Dwight Eisenhower that cover his service during World War I mistakenly indicate that he was responsible for organizing the 301st Tank Battalion at Camp Meade in March 1918. This is a minor error that partially originated with Eisenhower himself. In his book, At Ease: Stories I tell to Friends (Doubleday, 1967), on pages 136-137, Eisenhower indicated that he commanded 301st Battalion at Meade, a unit he expected to command overseas. However, Eisenhower himself wrote a brief history about Camp Meade shortly after the Armistice in which he recounted that the unit had initially been organized by Major Harry Stephens in February 1918, when the unit was actually designated 65th Engineer Battalion; it was not designated 301st Tank Battalion until spring, by which time Eisenhower had assumed command of Camp Colt, Gettysburg, Pennsylvania. (Assuming command of 65th Engineers when it arrived in Dorset, England, was fellow West Point graduate Major Henry E. Mitchell, son of Civil War General Winfield Scott Hancock’s aide-de-camp, William G. Mitchell.) This may be a somewhat pedantic observation, but this minor mistake has been repeated in pretty much every Dwight Eisenhower biography. It also appeared in a lot of newspaper summaries of Ike’s career published during WWII, so he may have just used 301st Battalion in his book to avoid confusion by introducing 65th Engineers into the mix.

    Related to the above, many like to assume that because “the veteran was there, so the veteran must have known what they were talking about.” Not necessarily so. I’ve encountered a number of instances in which men who served in the Tank Corps in WWI claimed to have served in, say, the “344th Tank Regiment” or the “1st Tank Division.” The US Army had no tank regiments in the First World War; there were companies, battalions, and brigades, but no regiments or divisions. (“344th Tank Regiment” should have read “345th Tank Battalion.”) Some members of the 301st Tank Battalion claimed to have launched offensive action against the Hindenburg Line from the Argonne Forest. However, 301st Battalion was attached to British 4th and 2nd Tank Brigades in British Fourth Army’s sector during its combat operations of September-October 1918. Not only was the unit deployed about 100 miles northwest of the Argonne Forest, it would have been quite impossible to get their heavy Mark V and Mark V* tanks into the forest. Indeed, it would have been foolish to attempt such a maneuver. Further, the Meuse-Argonne area was where American First Army operated, along with the aforementioned 344th and 345th Tank Battalions, both part of George Patton’s 1st (Provisional) Tank Brigade (later 304th Tank Brigade). But even Patton’s brigade didn’t attack into the forest; it operated along the eastern edge of it astride the Aire River.

    None of this is earth-shattering stuff, to be sure; but their cumulative effect is to unnecessarily distort what actually occurred. That distortion is avoidable much of the time by exercising due diligence, as well illustrated in your article, Greg. My mantra for conducting research, echoing Ronald Reagan’s refrain about missile limitations talks with the Soviet Union in the 1980s, is “Trust, but verify.” Without a time machine – and probably even with one – we’ll never know what “The Truth” was, but with careful consideration of sources at least we can get a little bit closer.

    1. Patrick, thank you for the thoughtful comments. I was hoping the blog would generate such comments from archivists and researchers about their experiences dealing with documentary evidence. Greg

Comments are closed.