Go back to article: Ventriloquised voices: the Science Museum and the Hartree Differential Analyser

The 'Mathematics and Computers' gallery, the Manchester Machine, and the National Physical Laboratory

Assistant Keeper Jane Pugh wrote her first letter about the Hartree Differential Analyser on 5 November 1973 to Jack Diamond (a Professor at the Simon Engineering Laboratory in Manchester). Through the lens of ventriloquism, her letter reflects the impact that Calvert had in changing the voices of the Trainbox, and the challenges that these new voices caused due to their incongruous relation with the object’s physical form.[29] Pugh explained to Diamond that a new gallery was in development at the Science Museum called ‘Mathematics and Computers’, for which she enclosed a list of potential objects and the gallery layout.

Figure 8

A list of items in the Mathematics and Computers gallery of the Science Museum

The list of items in the Mathematics and Computers gallery. Point 54 (C) is the ‘Trainbox,’ MS0237/23

Pugh wrote that she had been searching for more pieces for the exhibition on differential analysers. However, the only success she had had was with ‘Dr. [Arthur] Porter [who had] kindly given [them] the logbook of the first large Manchester machine, and we already have a Meccano model differential analyser made by Hartree and Porter’ (emphasis added).[30]

Figure 9

Drawing of the layout of the Mathematics and Computers gallery of the Science Museum

The layout of the Mathematics and Computers gallery. The cross in the bottom left is where the differential analyser exhibit sat, as part of the ‘Calculating Machines’ section, MS0237/23

The letter is fascinating in that it demonstrates both the strength of Calvert’s ventriloquism, and Pugh’s understanding of the incongruities it presented. On the one hand the language in the letter demonstrates the extent to which Calvert’s ventriloquism had changed the Museum’s understanding of the Trainbox; Pugh’s phrase ‘a Meccano model differential analyser made by Hartree and Porter’ indicates that she thought the Trainbox was the original object that had been created in 1934 by Hartree and Porter, and not the model rebuilt in 1947 by Hartree. Her use of the phrase ‘model differential analyser’ also implies that at the ‘hands-on’ level in the Museum, the voices of the original 1934 object had been successfully ventriloquised through the Trainbox, changing the Museum’s internal understanding.

On the other hand, the rest of Pugh’s letter hints at her knowledge that the object had a number of incongruities that meant it could not effectively tell the story of differential analysis as a result. Pugh describes her intention to make the exhibit on differential analysers in the Mathematics and Computers gallery ‘…the only one of its kind in the country...as complete as possible by addition of parts from the Manchester machine’.[31] The phrasing of her desire to make the gallery ‘as complete as possible’ through collecting other material implies that Pugh understood that the Trainbox was insufficient to tell the story. Her assertion that the exhibition would be ‘...immeasurably enhanced by the addition of parts from the Manchester machine’ implies that Pugh was aware that the Trainbox was not physically the same object as the 1934 model that had served as a proof-of-concept for the Manchester Machine and so could not be used to tell the stories of differential analysis in the same way that it had previously.[32]

Pugh’s other correspondence from this period indicates that her concerns about the provenance of the Trainbox began after discussions with Professor John Crank (a former student of Hartree’s) about the role the analysers played in the Second World War. Through their discussions, Crank explained the differences between Hartree’s and Porter’s original model and the ‘Trainbox’ that Hartree had created in 1947.[33] It was as a result of these discussions that Pugh contacted Arthur Porter to provide further context on his and Hartree’s work. Porter replied, sending Pugh a number of papers, notes, and letters related to his work on the 1934 object as well as the Metropolitan Vickers Logbook that detailed the construction of the Manchester Machine in 1935.[34] These sources described an analyser with both an input and output table that could resolve many differential equations and produce graphical outputs (Porter, 1934; Hartree and Porter, 1935; and Hartree, 1946). Clearly the Trainbox that Pugh observed in the Museum was much smaller and did not have input and output tables.  Pugh began to question the provenance of the Trainbox, visiting Jack Michel (another former student of Hartree’s) at the NPL to learn more about the work he had done with Hartree and about the use of the differential analyser during the war. In the reflective report she wrote after the visit, she described how Michel’s explanations of the original 1934 analyser helped to resolve the questions she had had about the challenging provenance of the Museum’s ‘small meccano d. a.’.[35] She recounts Michel’s explanation as follows:

‘After Hartree’s Meccano d.a. was broken up, Hartree retained some pieces which he had made up into a single integrator d.a. These could be easily packed up into a single case and transported for giving lectures. He used it for the problems of tractive force: as applied to railways.’[36]

Michel’s explanation made it clear to Pugh that the original analyser and the Trainbox were two distinct objects. This information was revelatory for Pugh, who wrote: ‘This seems to be a very likely explanation! I had heard of Hartree’s ‘train box’ from Prof. Crank, but had not connected it with our Model!’[37] This disconnect in Pugh’s understanding of the object demonstrates the success of Calvert’s ventriloquised voices in changing how the object was subsequently understood and exhibited at the Museum.

In contrast to Calvert’s correspondence, it is essential to understand that Pugh’s questions about the provenance of the Trainbox represent her attempt to revisit the voices ventriloquised through the Trainbox and amend them; the Museum’s success in collecting part of the Manchester Machine in 1974 meant that Pugh now had an alternative way of telling the story of differential analysis. Therefore, her work to determine the provenance of the Trainbox represented her attempts to ventriloquise new voices for the object to make sure they more accurately reflected the functions of the object itself as she now understood them. The acquisition of the Manchester Machine was vital in this process, and its importance was reflected in the handbook for the new Mathematics and Computers gallery. Here the differential analyser exhibit, which contained the part of the Manchester Machine and the Trainbox, was described as follows:

As can be seen from the University of Manchester Machine, rebuilt especially for this exhibition, early differential analysers were entirely mechanical, with integrators and other units interconnected by driving shafts.[38]

This label for the part of the Manchester Machine sat next to the Trainbox that was also on display. As the Trainbox was just a single-integrator from the original Hartree Differential Analyser with no input or output tables, this label would have highlighted the incongruities of the description attached to the Trainbox, of an object that was a ‘working model differential analyser’.

Figure 10

Pamphlet entitled a guide to computing then and now

The front cover of the handbook given to visitors to the Science Museum’s Mathematics and Computers gallery in 1974, a ‘Guide to Computing Then and Now’, MS0237/23

The handbook that accompanied the gallery contained very few details relating to the Trainbox object that was on display. This is not surprising given the content of Pugh’s reflective report, and her new understanding of the incongruous relationship between the previous ventriloquised voices and the physicality of the object itself, a demonstration model that only had a single integrator connected to no other units. The collection of the Manchester Machine allowed Pugh to tell the story of differential analysis and attempt to re-ventriloquise the voices of the Trainbox in the process. Yet Calvert’s ventriloquised voices – and the incongruities they caused – remained a part of the Trainbox. Despite Pugh’s best efforts to ventriloquise the object with a voice closer to the evidence she had found, the incongruous ventriloquised voices continued to dominate subsequent interpretations of the analyser throughout the next two decades, until the establishment of the Museum’s Model Walkway in the year 2000.

Component DOI: http://dx.doi.org/10.15180/181005/005