Collaboration and mediation: a guide to the creation of the Stephen Hawking Archive

Mediation, authorship and communication in the Archive

In her study of the work of Stephen Hawking with assistants, computers, in the media and in other forms of representation including the Archive, Hélène Mialet (2012) explored ideas and practices that mediated aspects of Hawking’s identity and legacy.^[2] In locating Hawking as the knowing subject of his works, Mialet found him represented in different media, distributed across writings, diagrams, interviews and artworks.

Studies in scientific authorship (Biagioli and Galison, 2003, p 6) note the fragmentation of authorship in the late twentieth century, for example in corporate, official or collective authorship in large collaborations. In contrast to some of these examples, Hawking tended to work in small groups and was a world-famous author. Thus, the process of curating the Stephen Hawking Archive is in some senses a process of locating the author through his work with amanuenses, typists and computers.

By the 1980s, archivists and historians noticed changes in working practices of scientists and associated changes in their recording media. Archivists working at MIT in the second half of the 1980s directed attention to this (Hass et al, 1985, p 23), leading to the development of new documentation strategies. The American Institute of Physics (1992–2001) brought together archivists with other disciplines to study multi-institutional collaborations, including their recordkeeping practices, in part to identify records of these collaborations, where they were located and by whom they were created and kept. Documentation strategies sought to identify and document important developments, structures, decisions and events, as well as the primary records and recordkeepers, so that important records could be targeted for retention to support future research.

Despite the differences in scale, Hawking’s office and network were involved in collaborations, in part following the practices of the scientific community he inhabited, in part necessitated by adaptations to new authorial practices because of Hawking’s ALS. Thus we have looked to Hawking’s working practices and processes of scientific communication to understand how records were created. This has informed the identification of records in the Archive and also a consideration of what is missing and where it might be located. In the following section we provide a very brief history of the Stephen Hawking Archive and reflect on issues that have arisen in the course of our work.

A brief history of the Stephen Hawking Archive

The Stephen Hawking Archive consists of correspondence and papers relating to Hawking’s scientific work, family and personal life. But in addition, there are elements relating to his political and social activism including the social responsibility of science,^[3] campaigns for accessibility, use of assistive technology, and involvement with public science and mainstream media, as well as other record formats (such as awards and certificates, gifts and commemorative items, ephemera and objects, photographs, and audiovisual material). A more detailed description is available in the online catalogue and further information will be added over time. There are a small number of digital media in the archive, including digital data tapes, floppy disks and optical discs. Labelling suggests some of these contain software, lecture presentations, photographs or video, whereas the content of others is not apparent from visual inspection.

In March 2021 the Stephen Hawking Archive was allocated to the University of Cambridge under the Acceptance in Lieu (AIL) scheme, which settles a proportion of inheritance tax in return for the donation of ‘important cultural, scientific, or historic objects and archives’ to the Nation (Arts Council England, 2023). The Archive was accepted by the University, received into the stewardship of the University Library, and accessioned. The Acceptance in Lieu process and the selection of material for the Archive had two effects on our work, which have led to some of the reflections in this paper. Firstly, although modern archival theory and practice in most European traditions envisages a role for archivists in appraisal and a strong connection between appraisal and acquisition (Couture, 2005), we were not involved in the appraisal and selection of material for the Archive. Secondly, the Archive for the most part did not include digital files. Similarly, the assistive communications technology artefacts included in the gift of Stephen Hawking’s office contents to the Science Museum as part of the same Acceptance in Lieu process did not contain data carriers. In separating data from objects, the latter might be considered what Kirschenbaum (2016, p 216) identifies as numinous objects, distinguished by their close association with a person, concretising their experiences and used to tell stories.

The Stephen Hawking Archive was selected by specialists from Christie’s auction house on behalf of the Hawking estate from a pool of material created and kept in Hawking’s University of Cambridge offices and residence. Most of this material accumulated in a storeroom in the University’s Betty and Gordon Moore Library between around 2001 until shortly before his death in March 2018. (In 2001 Hawking moved offices with the Department of Applied Mathematics and Theoretical Physics (DAMTP) from Silver Street to the Centre for Mathematical Sciences on Wilberforce Road, which the Moore Library is located alongside.) A smaller but still substantial tranche of material remained in Hawking’s office and those of his assistants until after his death.

The Archive was received and accessioned in March 2021, a process which involved checking items against the inventory provided in the Acceptance in Lieu process, adding missing information about additional items, checking the physical condition and re-boxing material into archival quality boxes and moving the collection to secure, environmentally controlled storage. In November, Susan Gordon started as the Project Archivist and surveyed the Archive, including detailed examination and identification of each folder or grouping of associated material or item, and the preparation of a more detailed listing than was previously available. The lack of a discernible original working order in most of the Archive resulted in its substantial rearrangement. This process involved identifying records with shared characteristics such as administrative function or subject and arranging them into ordered series. While further discussion is beyond the scope of this paper, archival arrangement is a central aspect of the archival mediation process.

Throughout this process, we have continued to think about appraisal, considering Hawking’s role in different endeavours and domains, achievements, life events and travels, identifying collaborations and establishing timelines for assistants, to relate all of these to the Archive. We have worked outwards from what has been included to consider what might have been created and retained and tried to understand the significance of what has been included in this broader context. We have also considered the potential significance of what has not been included in the Archive, for example most of Hawking’s digital files, and we remain open to the expansion of the Archive in future. We recognise that as existing parts of the Archive are curated, including small numbers of digital media, this will establish a framework for the integration of further material and documentation of the Archive’s interrelationship to material accessible at the Science Museum and elsewhere. We will also gain experience in archiving a small number of digital media and develop our capability to incorporate it on a larger scale. Now turning to the documents themselves we explore what the Archive can tell us about Hawking’s working world.

Stephen Hawking’s working world – theoretical cosmology, collaboration, amanuenses, writing and drawing in Hawking’s early career 1962–1973

Understanding the scientific and administrative contexts in which the Archive was created helps us understand the significance of distinct kinds of documents and their relationship not just to Hawking’s and others’ authorship but to an understanding of the ways in which he worked. In the late 1940s and early 1950s cosmology was a small, primarily theoretical discipline (Kragh, 2002). During Hawking’s time as a PhD student under the supervision of Dennis Sciama in DAMTP at the University of Cambridge (1962–1965), new mathematical approaches to cosmological problems and methods of analysis were developed and new phenomena observed. New communities of scientific practice converged around these approaches (Ellis, 2014, p 3; Robinson, 2019, pp 182–3, 230–1; Wright, 2014 pp 102–3). Participation in the community that formed around mathematical study of general relativity relied in part on the production of non-verbal symbolic communication using words, numbers, mathematical symbols and diagrams, and for Hawking it relied on working around his deteriorating handwriting. When read in the context of other sources, documents in the Archive provide evidence of how he produced drafts with amanuenses and collaboratively with other authors.

The wider context for these collaborations on paper were personal interactions, requiring mobility and elements of performance that were central to Hawking’s career and evident in the Archive. While the effects of Hawking’s medical condition presented obstacles to his personal mobility, in the context of a sociable, international discipline, he was remarkably mobile, travelling to numerous events including scientific meetings, lecture series and awards ceremonies. In the process of cataloguing the Archive, correspondence and other papers relating to 96 journeys outside the Cambridge area between 1967 and 2008 have been assembled into a series of files, with some of these trips having complex itineraries involving multiple events and locations. Evidence of travel is found in other series as well, such as photographs, gifts and commemorative items, and audiovisual recordings. Hawking is well known from biographical material and media coverage to have undertaken many more trips than this, including visits to other universities and learned societies. Like the seventeenth century natural philosophers discussed by Stevin Shapin (1990), Hawking was working in a collaborative world. He came to the attention of other scientists beyond his circle in Cambridge at a meeting at the Royal Society. Following a paper by cosmologist and proponent of the Steady State theory of the universe Fred Hoyle and his PhD student Jayant Narlikar, Hawking raised his hand and said that the calculations in the presentation were wrong, which Hawking knew because he had worked it out in his head (Hawking, 2000, p 60). He was developing the reputation of a daring postgraduate student who made his presence felt and who could do some calculations without pen and paper, although as discussed below, we see that ink and paper tools of one kind or another were sometimes indispensable.

The Archive contains a bibliography comprising 183 items authored or co-authored by Hawking until 2002, including technical and popular writings, books and articles arranged chronologically (not strictly by publication date, but possibly by submission date), and this^[4] formed the basis of the arrangement of a series in the Archive.^[5] A mixture of drafts, preprints, proofs and reprints/offprints of 108 of these publications, as well as additional and later publications, were found scattered throughout the Archive as it was received by CUL. Within this series, drafts dating from the early 1960s to early 1970s include a mix of handwritten drafts and typescripts by Hawking, some with corrections in his own or other hands. Sometimes there is more than one version including full or partial drafts in different formats. In another series^[6] there are early examples of notes, including mathematical expressions in Hawking’s own hand, such as 15 leaves of notes,^[7] in part relating to the section on manifolds in ‘Singularities and Geometry of Space-Time’. This was an essay submitted for the Adam’s Prize,^[8] one of the oldest and most prestigious prizes awarded by the University for the best essay by an early-career mathematician. Roger Penrose was awarded the prize in 1966 with additional prizes awarded to Jayant Narlikar and Hawking.^[9] Yet even as his scientific star was rising, Hawking’s handwriting declined.

Hawking commented that his choice of theoretical physics was fortunate because it is all in the mind (Hawking, 1988, p vii). Yet historical accounts of mathematical work by Andrew Warwick (2003) showed (in the case of nineteenth century Cambridge) that writing instruments are required and that the choice of tools affects the kinds of theoretical work that can be performed. This can be demonstrated in the context of Hawking’s work in the second half of the twentieth century as well (Wright, 2014, p 105). Theoretical cosmologists did have their own important, if humble, tools of pen, paper, blackboard and chalk for writing mathematical statements, diagrams, and scientific papers. From the late 1960s (as discussed below) some added computers to their toolkit for calculation and analysis but working with pen and paper appears to have continued in parallel in the discipline until the very end of Hawking’s career, for example in his work with Malcolm Perry and Andy Strominger on their ‘soft hair’ thesis (Galison, 2020). Examples from the Archive show how Hawking worked with pen and paper and drafted documents including scientific publications in the context of his deteriorating handwriting, often working with Jane Hawking as an amanuensis on documents such as his PhD thesis, a fellowship application to Caius College and the draft of a scientific paper.

A draft letter requesting a reference from Hermann Bondi for a Junior Research Fellowship at Caius College in Hawking’s hand shows uneven writing and blotches of ink.^[10] Hawking’s fiancé, Jane Wilde (m. 1965) wrote out his application to the college in longhand one weekend in February 1965 (Hawking, 2000, p 63). Before studying modern languages at Westfield College (1961–1964), Wilde took a secretarial course where she learned typing, which she did not enjoy but later used often (Hawking, 2000, p 22). Although Wilde (much to Hawking’s disappointment) was unable to type up his college fellowship application as she had injured her arm at a Westfield College bop, after marrying Hawking in July, she typed up his thesis, which he wrote in an illegible hand during the week and dictated to her on the weekends (Hawking, 2000, p 86). Jane undertook further typing, transcriptions and translations of Hawking’s work at different times throughout their married life of 25 years. There is evidence in and beyond the Archive that Hawking also used dictation to generate documents with paid assistants in university departments who later supported his work.

A later example of the deterioration of Hawking’s writing by the early 1970s is found in a long manuscript in the Archive.^[11] This is the text of a lecture for Les Houches Summer School in 1972 (Hawking, 1973). This annual summer school in theoretical physics was set up by Cécile DeWitt-Morette in 1951 to contribute to reconstruction of scientific institutions in France through international collaboration (Verschueren, 2019). The topic of the meeting in 1972 was black holes and Hawking recalled that ‘we solved most of the major problems in black hole theory’(Hawking, 2013, p 70). Hawking gave a paper on the event horizon, which is a boundary in spacetime beyond which events cannot affect the observer, referring to a black hole’s boundary and the boundary of an expanding universe. This manuscript draft is the last extended document in Hawking’s hand in the Archive. It is exceptionally long, containing 187 pages in Hawking’s hand plus additional pages in the hands of amanuenses, including that of Jane Hawking.

One example of a scientific collaboration documented by records generated using pen, paper, typewriter and printing is with Roger Penrose. Hawking first met Penrose in the context of efforts by he and his Cambridge colleagues to learn new methods developed by Penrose, which involved attending lectures and seminars in London and Cambridge (Ellis, 2014, pp 3–4). A memorable event for this network was the fourth International Conference on General Relativity and Gravitation at Imperial College London in July 1965, which notably included several Soviet scientists including Isaak Khalatnikov (Robinson, 2019, p 230) and focused attention on cosmological singularities. Khalatnikov’s work with Evgenii Lifshitz had ‘showed that the universe could have had a singularity – a big bang’ (Hawking, 1988, p 49). This problem formed the topic of Hawking’s early scientific publications and PhD thesis, in which he applied Penrose’s methods to the universe. The conclusion of this work was an important collaboration between Hawking and Penrose on a 1970 paper (Wright, 2014, p 100), in which they argued Einstein’s theory of general relativity showed the universe begins with the Big Bang and ends in black holes. They demonstrated that regular spacetime would exhibit singularities in the future and the past, pointing to the necessary formation of black holes in the universe. This paper was published in the prestigious Proceedings of the Royal Society (Hawking and Penrose, 1970). There are two typescript drafts in the Archive, one of which is annotated, and four copies of a diagram in two different sizes with manuscript captions and numbering in different hands.^[12]

These were Penrose diagrams that depicted unobservable mathematical objects and physical places like universes and black holes using conformal transformations, which distort the size of objects but preserve the angles between them like a Mercator Projection (Wright, 2014, p 109). This allowed points at infinity to be treated on the same basis as finite points, visualising and changing concepts of spacetime (Wright, 2014, pp 108–10). Penrose drew on precedents of Minkowski spacetime diagrams and ‘impossible objects’ – objects that present the visual effect of an impossible structure due to sudden changes in interpretation of spatial relations (Wright, 2013, p 137). Penrose developed these jointly with his father, the geneticist Lionel Penrose, inspired by artists like M C Escher. Wright (2014, p 105–6) situates Penrose diagrams in related contexts of research and pedagogy, explaining that they were not calculational or mnemonic devices. Mialet (2012, pp 65–78, p 71) describes how Hawking later in his career worked with students using Penrose diagrams, in a division of labour where students undertook pages of calculations, without which (White argues) Penrose diagrams cannot be analysed. This allowed Hawking to concentrate on the geometrical concepts that were represented by mathematical expressions: ‘I choose to concentrate on problems that can be given a geometrical, diagrammatic, interpretation… Often I work in collaboration with someone else, and that is a great help, because they can do all the equations’ (Harwood, 1983, p 19). The example of Penrose and Hawking’s 1970 paper in the Archive demonstrates how a particular method of mathematical analysis using pen, printing and paper to create diagrams enabled Hawking to collaborate with Penrose, and serves as a model for how Hawking worked with students later in his career.

The Archive provides further evidence of how Hawking worked with others, sharing ideas and texts. Hawking’s first collaboration with George Ellis, for example, ‘Singularities in homogeneous world models’ (Hawkins sic and Ellis, 1965), dates from the time Hawking was still a PhD student and Ellis (who was awarded his PhD in 1964) was a research fellow in the same department. This work resulted in two copies of a typescript in the Archive with approximately thirty algebraic terms added in Hawking’s hand. There are also letters from George Ellis to Hawking while Ellis held a visiting fellowship at the University of Texas, Austin in 1966. An aerogramme letter from Ellis to Hawking, postmarked 20 June 1966, outlines how they might collaborate, noting that it is difficult by post. It appears they considered collaborating on an essay for the Gravity Research Foundation prize although Hawking was more advanced in the topic and Ellis was working on other things. Two other undated letters from Ellis in the same file on University of Texas letterhead cover similar topics and appear to date from the same year, suggesting that they submitted separate entries. Hawking sent Ellis preprints, an outline and a manuscript and Ellis intended to send Hawking a preprint.^[13]

A later collaboration between the two men shows how they collaborated in the production and revision of a book and this in turn engaged a wider circle of colleagues, for example in reviews. The Large Scale Structure of Space Time (LSSST) (Hawking and Ellis, 1973) summarised geometry of four-dimensional spacetime in general relativity and some of its physical consequences, describing topological methods developed by Penrose and Hawking in the early 1970s. It was for specialists in this emerging field and was published in 1973 by Cambridge University Press. LSSST was largely written during a period when Ellis was a lecturer in DAMTP (1969–1974) and held several visiting appointments including in Hamburg in the summer of 1972, when the final text was being prepared and Hawking wrote to Ellis twice to coordinate the final alterations. At the time of writing the second letter Hawking had not yet had time to incorporate Ellis’ changes as Hawking was so pushed to get his lecture notes for Les Houches written (referred to above). He also notes he had rewritten the jacket blurb.^[14] Ellis’ contribution to this collaboration found in the Archive includes five leaves of hand-drawn diagrams. For the second printing of the book, there is a copy of a typescript containing his corrections and amendments dated May 1974, including mathematical expressions. Prior to the copy being produced, these have been annotated with ticks, crosses and crossings out to indicate whether the corrections and amendments were accepted. The file also includes two pages of handwritten corrections possibly in the hand of Roger Penrose,^[15] which Hawking requested following his review of the book, and some of which he incorporated.^[16]

By the mid-1970s, mathematical tools of pen and paper were becoming less accessible to Hawking, but he became known for his ability to manipulate geometric shapes in his mind and continued to collaborate with other authors in a sociable and distributed scientific community. Hawking also worked with amanuenses, initially Jane Hawking, to produce documents.

'Possum' – ability and sociability in correspondence

Correspondence in the Stephen Hawking Archive displays the same evolution from handwriting to typing by amanuenses seen in Hawking’s scientific writings. As well as the impact of his deteriorating handwriting, the effects of life events and administrative arrangements on Hawking are visible in the Archive. Around 1973, Hawking obtained access to a modified electric typewriter so he could in theory type documents including correspondence and scientific papers himself. From autumn 1974, he first obtained assistance from paid secretarial colleagues, including help with typing from dictation. We can observe changes in the voice of the correspondence representing a directness or indirectness in Hawking’s communication relating to the correspondence subject (for example, research in comparison to requests for public engagements), which appears to change over time. We can also note what is missing from the correspondence.

From the early modern period until the later decades of the twentieth century, paper correspondence was an important element of scientific communication used to connect distant people who met rarely if ever, to collect scientific data and intelligence, and to arrange meetings for international travellers. Its purpose and forms changed significantly in that time – functions previously served by correspondence still existed but move from conventional correspondence to electronic forms of communication in the age of the internet, online publishing, social media and computerised data gathering and analysis. Here we provide a chronological review of the correspondence in the Archive, providing examples of content, and form, for example handwritten, typed, facsimile or word processed.

The earliest piece of correspondence in the Archive is part of a group of records retained by Hawking’s mother Isobel Hawking and eventually passed down to her son. This includes letters sent to Hawking in the late 1950s when his immediate family temporarily relocated to India for his father’s work, but he stayed behind in England for his schooling. It also includes congratulatory correspondence regarding being awarded a first-class degree at Oxford. The only piece of correspondence written by Hawking that predates his time at Cambridge is a letter he wrote in 1948 at age six, to his father who was working abroad.^[17]

There is small amount of correspondence from the mid- to late 1960s, during Hawking’s postgraduate study and early career at Cambridge. This is mostly incoming and outgoing correspondence with colleagues and collaborators, but there is also correspondence relating to employment, events and personal letters. During the 1960s there are examples of draft correspondence in Hawking’s hand, including drafts for telegrams and letters sent to John Wheeler and Physical Review Letters editor S A Goudsmit, in response to a misunderstanding that arose with Wheeler’s then student Robert Geroch.^[18]  The incident with Geroch has been reported previously using the other side of the correspondence in the John Wheeler Papers at the American Philosophical Society (Seife, 2021, pp 303–4). The latest example of draft correspondence in autograph is a 1973 letter regarding arrangements for a visit to the USSR.^[19] 

Noticeably more outgoing correspondence than incoming correspondence was retained from the early 1970s, resulting in fragmentary or one-sided exchanges. Almost all correspondence outgoing from Cambridge during this period is in the form of typed carbons, and it tends to be very succinct. An example of this is Hawking’s letter to Nature to accompany his paper ‘Black Hole Explosions’ on what would later become known as Hawking Radiation.^[20] It is unclear who typed the correspondence and none of it has a typist notation. At this point it would have been difficult for Hawking to type himself and he did not yet have a paid personal assistant (PA). It is unclear why there was an issue with retention of correspondence during this period, but the fact that Hawking was splitting his time between the Institute of Astronomy (IoA) and DAMTP may have affected recordkeeping,^[21] or records may have been lost in a series of moves including his departure from the IoA, departure from Cambridge to Pasadena for a year in 1974, and a house move upon his return to Cambridge in 1975.

In May 1973, Hawking wrote to a firm in Aylesbury, Buckinghamshire called Possum Controls Limited, referring to a demonstration of their controls for electric typewriters that offered the potential for typing speeds of up to 40 words per minute.^[22] The firm, founded by inventor Reginald Maling in 1961 with the name derived from the acronym Patient Operated Selector Mechanism expanded to the Latin possum – ‘I am able’ – supplied the typewriters with controls that could be modified for individual requirements. (Kirschenbaum, 2016, p 16 notes that the typewriter was initially conceived and marketed as an assistive technology.) Hawking requested a tandem typewriter with a stepping mechanism to raise or lower the line by ½ space increments, and to maintain the same position in the line when switching between typewriters.^[23] There is no example of this design in the Archive, but one may imagine the ‘tandem’ had letter and number keys and mathematical symbols, enabling the user to insert the latter into texts, so perhaps Hawking intended to type his own scientific papers. Hawking requested a quote for a personally modified system and loan of a standard 8 A system with a single joystick.^[24]

Hawking spent the 1974–1975 academic year at Caltech, and during this time was assigned a secretary, to whom he dictated correspondence and other documents (The Tampa Tribune, 1979). This period marks an increase in the amount of correspondence in the archive. Correspondence with members of DAMTP regarding matters such as staffing and event planning dramatically increased. Presumably this correspondence replaced communication that would normally have been done face to face and indeed it reduced following Hawking’s return to Cambridge.

Back in Cambridge, Hawking was appointed a Reader, and for the first time had a PA in DAMTP, (Hawking, 2007, p 241). The PA could help him generate correspondence and may also have filed and retained more correspondence. The volume of correspondence in the Archive dating from this period increases, as does the variety of correspondence. There is more personal correspondence, such as letters relating to health care and financial matters, and more correspondence regarding travel. Additionally, outgoing correspondence tends to become longer and more effusive, possibly as the burden of typing is now on someone for whom it is a key part of their role. Moving into the late 1970s the amount of correspondence relating to motor neurone disease (MND) and accessibility also increases. This may reflect the increasing impact of ALS on Hawking’s work and life, the increasing assistance with or retention of correspondence, or the intersection of these factors.

When Hawking wrote to Possum Controls Limited again in August 1975, he was about to return to Cambridge from California and asked for a progress update on his modified typewriter.^[25] A December letter of the same year asked when it would be ready and noted that it had been 2½ years since its construction had commenced. Hawking sought further information from Grange electronics about the programmable word and calculating facilities used with Possum typewriters, reflecting that the word store might be useful but that he would be unlikely to use the calculator. This was designed for accounting functions and was presumably too basic for Hawking’s needs. He requested the loan of another ordinary Possum typewriter again.^[26] In January 1977, Hawking’s PA at DAMTP wrote to request that Possum collect the presumably loaned typewriter and demonstrate their book stand and page turner.^[27] The Archive does not confirm if Hawking’s personally modified typewriter was built or delivered, however he does appear to have had access to a model 8A Possum control typewriter both before he went to California in 1974–1975 and following his return to Cambridge in 1975. This is confirmed in a letter dated 11 November 1975 in which he explained the operation of his special typewriter with two levers in eight positions, providing the correspondent with details for Possum Controls Ltd., although Hawking made no mention of how often he used the typewriter.^[28] A later press article (Harwood, 1983) indicated that he did acquire a page turner.

While the change of volume in some types of correspondence can possibly be explained by an increased capacity for creation and retention, Hawking’s career trajectory was also a factor, so analysis of patterns in the Archive both reflect and provide a map of Hawking’s life and work. An increase in correspondence regarding his receipt of awards and honours follows on from his scientific achievements. The amount of unsolicited correspondence or ‘fan mail’ increases in the late 1970s following Hawking’s early forays into popular science including appearing in the BBC’s Key to the Universe and writing an article for Scientific American in 1977 (Hawking, 1977).

World events also influenced the composition of correspondence. The increase in hostility between the USSR and the West during a late phase of the Cold War is marked by the appearance of correspondence both with groups such as Scientists Against Nuclear Arms, and relating to local, grass roots efforts regarding the Campaign for Nuclear Disarmament. Hawking’s correspondence also reflects the challenges of scientific collaboration with Soviet colleagues during that era, for example a letter from Martin Rees to Hawking in August 1978 requested the names of any scientists Hawking would like to invite to the 1979 or 1980 quantum gravity workshops so they could be discussed during a visit of the President of the Royal Society to the USSR in September.^[29]

There is a reduction in volume of outgoing correspondence reflecting an extended period of illness and then reduced ability to communicate following Hawking’s tracheostomy in late summer 1985. Correspondence in the following months includes responses to invitations generated by his PA that downplay the seriousness of his communication difficulties.

With the adoption of the Words+ Equalizer programme (discussed more fully below), Hawking regained the ability to generate text without the need for dictation. This is reflected in a printout of a letter to his parents written in early 1986,^[30] found together with other mementos of his life and work collected by his mother Isobel and eventually passed down to him. Interestingly, this is the only obvious example of hard copy correspondence clearly generated via the Words+ system found in the Archive. It is possible that Hawking did not habitually print out paper copies for filing, or have anyone do so, and within a few years communication by email would have been more feasible.

In the mid-1980s Hawking’s engagement of a literary agent, Al Zuckerman of Writer’s House, added another layer of complexity to his communication. Some correspondence that would previously have been sent to Hawking may have been directed to Zuckerman instead, and from the early 1990s onwards Writer’s House is a main correspondent. Much of this is by facsimile (fax); this may reflect later uptake of email in the industry, or the need to forward documents.^[31] Despite the success of A Brief History of Time (Hawking, 1988) the amount of unsolicited correspondence from the public, or ‘fan mail’ found in the archive decreases after the publication. According to biographical accounts the amount of fan mail received by Hawking’s office was so voluminous that it was not practicable (or considered important enough) to keep in its entirety (Ferguson, 2012, p 267) (White and Gribbin, 1992, p 270). This is a good example of correspondence that was created but not retained.

The early 1990s correspondence is dominated by communication relating to Hawking’s work in popular science publishing, with Writer’s House and Bantam, and about other popular science projects including the film of A Brief History of Time (1991). There is a notable shift in voice, with his PA communicating on his behalf referring to Hawking in the third person.^[32] This contrasts with earlier stages of his career, when communications, including those about popular science projects, were usually written in the first person, even if typed by a PA. It is unclear whether this reflects the practice of his PA at the time (Hawking had at least eleven PAs at DAMTP between 1975–2018), increasing volumes of correspondence requiring delegation to deal with, or a decision on Hawking’s part to put a degree of separation between himself and some aspects of his career. Communication pertaining to his academic life during this period is very limited in both volume and scope within the largely analogue archive, possibly reflecting earlier uptake of email in the academic arena; those examples that do exist are written in the first person. It is unfortunate that there is so little academic correspondence from this period for comparison. The absence of email in the Archive is an example of exclusion of most of Hawking’s digital archive from the Acceptance in Lieu gift, but this example shows how important it might be to providing an account of Hawking’s scientific collaborations in the later part of his career. Hawking appeared not to habitually print out emails with academic correspondents, although a limited number of emails related to contracts and in subject files were printed out by an assistant. The 2000s see the appearance of correspondence by Elaine, Hawking’s second wife (m. 1995) about communication technology, travel and other practical topics, probably brought from their residence. Hawking resumed his correspondence with Possum Controls as he appeared in their newsletter in 2000 as the customer for a Freeway environmental controller (a remote control that could be programmed to connect to other electrical devices around the home including beds, alarms, communication devices, radio, television, etc.). Intriguingly, correspondence about this article includes a reference to the Hawking’s ‘Museum’, for which an old Possum typewriter is sought.^[33]

A close study of Hawking’s correspondence in the Archive shows how he used correspondence to manage his scientific and non-scientific affairs, and how it was produced with human and technological assistance. The initial analysis here suggests that the overall pattern of correspondence in the Archive reflects Hawking’s life events and decisions about what to retain and include in the Archive, as well as reflecting different aspects of his work, and world events.

‘Can you hear me?’ – keeping Hawking talking

Hawking’s lectures, alongside his scientific writings, are an important expression of his ideas and how he communicated them. For much of his career, audiences unfamiliar with Hawking’s speech had difficulty understanding it easily, as it became less distinct and increasingly faint over time. Lectures were sometimes presented on his behalf by Hawking’s colleagues, or by Hawking himself with a student or colleague repeating his words. It was only after Hawking obtained computerised voice technology, discussed in the next section, that his speech could once more be clearly understood by the audience. Lecture slides and notes are the most direct form of evidence of Hawking’s ‘voice’ in the Archive (although recordings of lectures given after his tracheostomy are present in the audiovisual series, the content of which has not yet been examined). Most of the lecture material created in a variety of formats has been brought together into one series in the Archive. This was informed by how different groupings of lecture materials were received by the Library (see below on the acquisition and appraisal of the Archive) but also because delivering seminars and lectures to both research and public audiences was an important activity performed by Hawking (even if often voiced by others). There are also significant gaps in lecture materials in the Archive, which we also note below.

Early lecture materials include two sets of transparencies mounted in 3.25 x 4 inch slide casings. These are accompanied by some of the original line drawings which were photographed to produce the slides. One set appears to contain slides for multiple lectures. Alongside other notes and early writings, there are also a few sets of pages of brief handwritten notes written circa the early 1970s that appear to refer to slides, but these do not seem to correspond to any early lecture aids contained within the Archive and are not expansive enough to have allowed for identification of the lecture to which they refer.

Overhead transparencies circa mid to late 1970s arrived at the University Library grouped together within the Archive, mostly in labelled packaging. These are written in mostly unidentified hands, which vary between sets of transparencies and over time. Don Page’s handwriting is recognisable on two sets of transparencies from around the time when he was part of Hawking’s group in the mid-late 1970s, so it is possible that the others were written by postdoctoral researchers or students assisting Hawking with the presentation in question. Page was a PhD student of Kip Thorne at Caltech when Hawking visited there in the academic year of 1974–1975 and joined Hawking’s group at DAMTP as a postdoctoral researcher in autumn 1976, moving in with the Hawking family. The arrangement whereby one or more of Hawking’s students provided both domestic and research assistance started when students accompanied the Hawking family to Caltech and continued until the mid-1980s. At this point domestic assistance was augmented and eventually replaced by nursing care, after which students primarily worked with Hawking on research. There are small samples of the handwriting of some of Hawking’s students and postdoctoral researchers within the Archive, but no attempt has yet been made to try to match these to the handwriting on the transparencies. Some transparencies are accompanied by handwritten drafts of the transparency’s content. For example, for a talk ‘Black Holes and Unpredictability’ on the information paradox the transparencies are accompanied by a manuscript template in Page’s hand. The archive also contains a preprint of the version published in Annals of the New York Academy of Sciences, showing the relationship between speaking, writing and publication.

There is a gap in the lecture series from the late 1970s to 1987, except for a single set of transparencies from the Shelter Island (New York State) II conference in 1983, which were found separate from any of the other lecture aids. Other records of this type must have been generated during this time as Hawking lectured frequently. There are typescripts for published versions of talks given during this time within the published works series; it is only the visual aids that are missing.

The lecture series takes up again in 1987 with a group of files labelled with a title and often a year. Despite covering the period from 1987 to 2012, these were all received in the same type of hanging file cabinet folders labelled in the same hand, as though someone decided to organise (or at least re-folder) the material at some point. Assisting with lectures was a function of the graduate assistant (GA), but this sub-series begins after the genesis of this role in 1983. GAs were employed by the Department to assist with Hawking’s travel and lecture commitments and public inquiries, increasingly specialising in assistive communications technology. The type of lecture aids included in the files transition from overhead transparencies to 35mm slides, to a combination of the two, to paper printouts of presentations presumably created using the Microsoft PowerPoint application. Files sometimes contain multiple versions of lecture aids as lectures were often given on multiple occasions, sometimes with a different title. Some of the later files contain only a lecture script(s), without any visual aids, which hopefully still exist elsewhere in digital format. The lecture scripts themselves are often ‘marked up’ to indicate when to move to the next transparency/slide, and some contain intentional spelling errors to ensure Hawking’s speech synthesizer correctly rendered the pronunciation of words. Some files also contain additional versions of lectures that appeared to be formatted for distribution (with conventional spelling); these are sometimes duplicated in the published works series, if applicable. The lecture series seems quite comprehensive from 1987 until about 2000, at which point there is another gap, and then a single file titled ‘Public Lectures’ which contains scripts and occasional PowerPoint printouts for both public and technical lectures, dating from approximately 2007–2012. One standout within this file is a presentation called ‘Keeping Hawking Talking’ by GA Sam Blackburn, which outlines the history of Hawking’s use of assistive technology,^[34] discussed in the following section. All the other scripts are written in Hawking’s voice, often beginning with his characteristic ‘Can you hear me?’.

Additional lecture scripts from the late 1980s to the 2000s were found amongst the loose typescripts in the Archive. Those of these from the 1980s tend to be more biographical or focused on Hawking’s experience of ALS, whereas the later scripts are more varied, including popular science and technical lectures. These loose scripts have been assembled into files by the archivist and added to the end of the pre-existing series of labelled files. The end of the presence of lecture material in the Archive appears to be 2012, although as Hawking is known to have lectured after that date, it is hoped that additional records survive in digital format.

The Archive shows how lecture materials were produced in different media, using different technologies. While in earlier sections we showed how Hawking worked with collaborators and informal and formal secretarial assistants to produce scientific papers and correspondence, he worked with students, colleagues and graduate assistants to produce lecture materials. Part of the initial work on the Archive has been to gather different groupings of lecture materials and scripts scattered throughout the Archive to form the collection of lecture series. Our initial study of this shows the shift from indirectly addressing the audience through colleagues to directly addressing the audience using assistive communications technology. Assistive communications technology was one method used to produce correspondence and lecture materials. In the following section we focus on the development of Hawking’s adoption of assistive communications technology and place it in wider contexts, to illustrate its relevance for and impact on the Archive.

Computer-assisted communications technology (1977–2006) and the Archive

The Archive shows Hawking’s adoption of computerised assistive communications technology in the production of documents and evidences his early attempts to use a computer in his department at the University of Cambridge and at home. While secretarial and graduate assistants and students continued to produce certain kinds of documents in collaboration with Hawking and on his behalf (from administrative correspondence to lecture materials, mathematical calculations and diagrams) Hawking developed more agency for direct communication of speech and writing when he chose. In discussing Hawking’s early use of assistive communications technology, we place this in two important contexts. The first is the adoption of computers for calculation and analysis in Hawking’s field of general relativity and gravitation research. The second is the adoption of word processing in literary culture, which is relevant as in this period Hawking published his bestselling book on scientific work in his field – A Brief History of Time (1988). His work in this respect was comparable with other popular authors. As observed by literary scholar Matthew Kirschenbaum (2016), word processing was conceived and initially implemented to produce paper artefacts, not digital files, which were conceived as instrumental intermediaries between the author and their goal of clean copy, but not as persistent artefacts in themselves. Except for the few digital carriers with yet unknown content previously mentioned, the impact of the computerised production of documents on the Archive is primarily in the form of the paper artefacts generated. These considerations point to the importance to the Archive of the absent digital files. We conclude with the example of the genesis and production of A Brief History of Time, which allows a reflection on the paper artefacts of this process (which are present in the Archive) and some of their limitations for scholarship.

At the end of July 1985, Hawking flew to Geneva accompanied by students and nurses to take part in discussions with particle physicists at the European Centre for Nuclear Research (CERN) relating to Hawking’s work on the arrow of time (Hawking, 2000, pp 417 and 420). Already in that year Hawking had travelled for work to California, Chicago, Japan, China and Moscow as well as the Rutherford Appleton Laboratory at Abingdon for a conference on the Very Early Universe. During the first week in Geneva Hawking was diagnosed with pneumonia and placed on life support for treatment. He was transferred to Addenbrookes hospital in Cambridge, where further treatment included a tracheostomy operation, which removed his voice.^[35] The story of how Hawking regained his voice with a personal computer, software to operate it and compose text and a speech synthesizer is told elsewhere (Hawking, 2000, p 445, 453–4; Madeiros, 2015; Wozniak, 2019). Hawking outlined the elements of his new communication system in a press statement made in conjunction with the publication of A Brief History of Time in 1988, as media interest increased in his work and the way he did it.^[36]

More than a speech machine, the set-up could be used to write mathematical papers including mathematical notation using a formatting programme called TeX. This allowed selection of words that sounded out the equations that were then converted to symbols and printed out. Hawking could also write speeches, save them to disk and then recall and ‘speak’ them using his synthesizer, saving up to half an hour of speech in a single file, with the option to use more than one file. He could then answer questions at the end using real-time text generation and speech synthesis, although this was slower than delivering the speech. The speech synthesizer, a SpeechPlus CallText 5000, could also be used to dial and answer calls.^[37] The synthesized voice was modelled on the natural voice of speech synthesis researcher, Dennis Klatt (Wozniak, 2019, p 6). Having lost his ability to hand-write and speak, Hawking’s new communication system augmented his capability for and control over the generation of texts, mathematical notation and speech (conversation in-person, telephone and lectures sent from a prepared text), some of which would eventually be added to his Archive as printed artefacts.

Like others in office contexts Hawking had previously used dictation to create texts. Yet in an interesting reversal he now created text to generate speech as well as written documents. From the advent of advanced electronic typewriters discussed by Kirschenbaum (2016, pp 176–182), which recorded text on magnetic tape for correction and later printing, these opened new possibilities for authors to draft, print and revise text. For some users such as US author Paul Elkin, who was diagnosed with multiple sclerosis in 1970, word processing was itself an assistive technology (as his writing declined and typing was painful). The lighter touch of a keyboard was effortless by comparison to writing and Elkin changed his creative practice accordingly, using search mode to find thematic keywords and tighten plots. Like Hawking – some elements of whose communications technology were sponsored by firms that created them – Elkin was helped by colleagues to source the new technology. Washington University, where he was a writing professor, paid USD 10,000 for a Lexitron VT in 1979 on which he composed the second half of George Mills (1982) and later revised the entire novel (Kirschenbaum, pp 164–5).

Hawking’s medical crisis of 1985 did not mark a complete break with the past so much as urgently reframe his earlier search for assistive technologies to support communication and scientific authorship. His previous efforts to source a modified controls electric typewriter were discussed earlier in this paper where we noted that his second loaned modified controls electric typewriter was returned to Possum Controls Limited in 1977. In November 1977 quantum physicist John Polkinghorne, Acting Head of DAMTP, wrote to the Council of the School of Physical Sciences at the University of Cambridge requesting funds to set up a terminal controlled by a joystick. This was developed by the University Computer Laboratory connected to the IBM 370/165, one of the University’s shared computers maintained by the Computer Laboratory. The VDU would connect to the computer via the Department’s Post Office line. This would enable ‘Professor Steve Hawking to manipulate symbols on a visual display (including erasures and recall of past material) in a way that would approximate what one can do normally working on paper. Obviously, this would be a great advantage for Steve and would be greatly superior to anything one could achieve via electric typewriters, for instance (which would not have erasure or recall).’^[38] The total cost would be £1,250 with a small sum required for maintenance of the VDU. As described by Hawking to a correspondent in February 1980, in this arrangement he used two eight-position joysticks and a foot switch to operate the shift button, and, linked to the University computer by telephone line, he could input notes, file, and send them to a printer near his office. However, it was a slow process, so he preferred dictating material to his secretary.^[39]

As well as seeking access to computer technology at work, the Hawking household had a home computer. There is product literature relating to personal computers in the Hawking Archive from retailers in Pasadena, California dating from 1977.^[40] By July 1979 Hawking had purchased a Commodore PET 16K series 2001 computer. He also purchased software, including games, suggesting possible use by other members of the family including his son, Robert. In February 1981 Hawking wrote to the Commodore Information Centre in London, outlining his requirements as a mathematical physicist working on general relativity, cosmology and black holes but who due to ALS found it impossible to use a pen or typewriter keyboard, requiring controls to input text or programs. Although a friend had arranged for two joystick switches to be connected to the keyboard, as well as their mechanical deficiencies there were too many positions to control effectively whereas two four-position joysticks would allow the selection from 256 alternatives including letters and common words. Hawking contemplated increasing the memory or acquiring an additional disk drive, and the inadequacies of his old teleprinter that was slow and only printed upper case letters.^[41] It sounds like Hawking intended to use his home computer for drafting and printing text, and he mentions programmes for this, but mathematical notation is not specified.

Since 1979, when Polkinghorne first succeeded in linking Hawking to the University’s computer, scientific computing capacity in Hawking’s department had developed further. In February 1984 Hawking wrote to Cambridge-based technology entrepreneur Clive Sinclair explaining that the Relativity Research Group had recently been donated a new Universe computer by the Charles River Data Corporation of Boston, Massachusetts, a 68000 based machine with a megabyte of RAM and 35 megabyte of hard disk. It would be used for numerical calculation and mathematical analysis of black holes and quantum gravity, and word processing. They planned to connect the Universe computer to a printer and several personal scientific workstations with graphic displays. Thinking about using BBC or IBM personal computers for this purpose, Hawking now wondered if Sinclair would like to donate four or five of his newly released Quantum Leap computers to the Relativity Group: ‘In this way Quantum Leap would be making a big contribution to quantum gravity.’ Hawking pointed out the publicity opportunities this would bring to Sinclair’s firm, referring to the media that Hawking attracted, including the Horizon programme of 1983 and an article in the current month’s Reader’s Digest. Hawking was attracted to the Quantum Leap by its high-resolution graphic displays in colour and compatibility with the Universe computer enabling the easy transfer of assembler language programmes from the Universe including visualisation software. Hawking even wondered if Sinclair would be interested in some of the software developed by the General Relativity and Gravitation group, which might be used in other computing applications.^[42]

Hawking got his terminal with modified controls for the University’s IBM 370/165, but it is not recorded in the Archive if Commodore suggested or supplied modifications to his home computer, or whether Sinclair donated Quantum Leaps in the cause of quantum gravity. What is evidenced is that from 1977 Hawking was exploring opportunities to support his group’s work and individual assistive technology needs with innovations in three centres of development for commercial computer technology – Cambridge in the UK, and California and Boston in the US – and that he was interested in computer applications for word processing, mathematical calculation and analysis, and visualisation. How the daily need for his office to keep up with increasing amounts of correspondence and a steady stream of publications was met can be inferred from patterns in the Archive. From 1983, most of Hawking’s scientific papers there were printed out from computer generated files rather than typed, suggesting the adoption of computers for drafting papers from around that date. Much of the correspondence originating from his office appears to have continued to have been typed until at least 1985, suggesting the later adoption of word processing for generating correspondence by Hawking’s personal assistants.

Hawking’s press release about his communications technology was issued in the context of the release of his popular science book, A Brief History of Time, which was drafted and edited in the period before and after the setup described in the release was obtained. We do not cover this in detail, as what remains in the Archive are the paper artefacts of a computerised document creation process. This has limitations for the study of Hawking’s writing of A Brief History of Time because while it documents the end products of this process, and provides snapshots of the drafting process, it does not replicate a completely manual drafting process that potentially might be preserved in its entirety. We suggest that further analysis (for instance referral to publishers‘ archives, and additional oral or written sources) may be fruitful in understanding the process in developing this book.

Digital archiving in a socio-technical system – graduate assistants’ notebooks (2006–2018)

As described above our initial analysis of the Archive shows that Hawking had the means to create digital files from the late 1970s, and started creating them by necessity after 1985, in the context of the development of assistive communications technology at work and at home, although the outputs of these digitally created files are represented in the Archive in the form of paper artefacts. But what about subsequent developments in Hawking’s assistive communications? Here we use paper sources in the Archive and beyond to show that the preparation of a digital archive was considered by Hawking and his assistants. Digital files and media were retained and assessed with a view to potentially including them in a digital archive in a public repository to preserve Hawking‘s legacy and to support future research. We briefly discuss digital archives, and how the management of digital archives may differ from the management of primarily paper archives. It is interesting here to reflect on possible reasons why the majority of Hawking’s digital archive was not included in the Archive offered under the Acceptance in Lieu scheme, and we consider some of the challenges to the acquisition, preservation and access to digital archives, and how some of these might be overcome. In general, we argue that the digital files generated by Hawking and his assistants should be added to the Archive in future, with the Archive documented thus far providing a framework into which digital files might be integrated or compared with paper artefacts of digitally created files and other records. In this case an understanding of the development of Hawking’s assistive communications technology provides important background for more recent attempts to consider his digital archive.

Since 1988, when Hawking issued a press release about his communication system, further improvements and adaptations occurred. From 1997, Hawking’s computer-based communication system was supported by Intel® Corporation and resulted in several developments. These included new laptop computers (Hawking, 1997), an infrared cheek sensor attached to Hawking’s eyeglasses to stop the cursor and select a character (initially developed in 2008 to replace a finger-operated switch) and replacement of his navigational and textual composition software in 2011. By 2018 Hawking had two complete wheelchair systems with custom-built communication hardware and software,^[43] recalling the multifunction purpose of the astronomer’s chair as documented by Omar Nassim (2021). The only element that was not replaced until shortly before Hawking’s death in March 2018 was his hardware voice synthesizer, an emulation of which was accepted by Hawking in January that year (de Lange, 2011; Wozniak, 2019).

As Mialet (2012, p 169) has shown, Hawking’s office established a socio-technical network to support mobility, communication, text and other symbolic production required by a twenty-first century theoretical physicist and public figure. Hawking developed scientific theories based on mathematical analyses, spoke, wrote, found information, met with colleagues, responded to interviews and went to conferences. Based on the developments in his assistive communications technology and his description of how he used them, we would expect that much of this activity was reflected in his digital files. This system was constantly evolving, drawing new collectives into the work of maintenance and innovation required to keep it going. The assistive technology activity of Hawking’s office brought about increasing specialisation, so that from around 1983 the increasing complexity of his communication and travel requirements was a key driver for the emergence of the new role of graduate assistant (GA). While Hawking’s postgraduate students helped with research including shaping research, mathematical work and writing papers, the GAs looked after the technology, travel arrangements and lecture preparations, eventually answering or redirecting emails from members of the public. Hawking continued to work with his office using a combination of in-person communication assisted by technology, so both personal and technological mediation remained important in the formation of the Archive.

The GAs’ notebooks in the period 2006–2017 provide some insight into this process. Their notes, diagrams and drawings depict the complexity, maintenance and innovation behind Hawking’s assistive communications technology, as well as how Hawking communicated with assistants and how they thought about managing the textual and voice outputs of this system.

In 2009, one GA documented a special project to scope the management of Hawking’s digital archive, following initial efforts to gather Hawking’s paper archive. Goals identified for such an archive included an approved complete Corpus and Archive. Identified risks included loss of important data for future publication, or the inadvertent publication of private information or information taken out of context as well as potential media interest in decisions about Hawking’s digital archive.

The GA consulted members of Hawking’s circle, who each provided different instructions or advice. The notes go on to identify a range of media which may be relevant, whether files are present, storage capacity and last modified dates, as well as referring to email accounts. On 28 April 2009, a further comment from Hawking was recorded.

Stephen: I had a dream which made me tell [the GA].

My documents [should go on the master copy].

Everything that looks important [should go on to the master copy].

Don’t destroy anything yet.

Leave emails alone.^[44]

The notebooks are fragmentary works in progress, aide memoirs for a busy GA to identify and record tasks and conversations and to capture ideas. Detailed travel arrangements are juxtaposed with wiring diagrams, sketches for technical improvements, and occasional comments by Hawking recorded by the GA. These fragments of conversation cannot be regarded as decisions or definitive wishes, yet they indicate that a digital archive was at one time on Hawking’s mind, and his office commenced the process of exploring this question. Hawking, the GA, and others who offered their views articulated a concept of a digital archive, what its purpose might be, what kinds of things it might contain (e.g. lectures) and where it might be found. These notes indicate the ubiquity of data storage devices, media and files and their potential relevance as sources of a digital archive, although they do not record the selection of files to be included.

One other source that provides a clue to the contents of Hawking’s digital files around this time is a comment reported in Mialet’s study (2012, p 169) from an interview with a GA circa 2007. This refers to an Equalizer folder containing plain text files of everything important to Hawking. (Equalizer was a communications software used to select words that can be sent to a speech synthesizer.) If this folder or other digital files still exist, they are important potential sources for extending the scope of the Archive, as the amount of printed correspondence and papers dating from the early 1990s onwards is significantly less than that retained for earlier periods. This suggests a shift in Hawking‘s office away from printing documents out to share and retain them and towards sharing and retaining them on digital media.

When Mialet (2012, pp 165–171) discussed the Archive in the context of Stephen Hawking’s work, his digital archive was considered an important disruption to previous concepts of the Archive and systems of archiving. Some of Mialet’s questions concerned how digital archives could be preserved and many of these questions have been addressed by the archival and digital preservation communities. Efforts have focused on how to preserve characteristics valued in physical archival material such as uniqueness and inalterability, provenance and original order, through preserving computer and human generated metadata (descriptive or contextual information) associated with the generation, transfer, processing and storage of files. Numerous international and national standards and guidance exist to help archivists and others define and preserve these characteristics in digital files.

Mialet (2012, p 139) also discussed what should be preserved; in other words, questions of archival appraisal. Appraisal of electronic archives has focused on defining the requirements to create, retain and use electronic records to document activities of organisations, groups and individuals, through the up-front definition of what needs to be retained and the proactive organisation and management of files through time. But the focus has been on retaining the same kinds of files or ‘evidence’ as would have been retained in paper archives of an earlier era, including single copies of important documents generated by their first or authorised creator or by the person or group responsible for creating them. Digital formats such as email have largely been studied in terms of how to preserve them – not in terms of how that might change what should be preserved, although the practicalities of preserving digital archives do often lead to different appraisal decisions (for example, the decision to preserve an entire email inbox of an archive creator that includes their work and personal email and important and trivial correspondence, because it is infeasible to select individual emails). In appraising digital files, archivists still tend to prioritise the ‘original’ or ‘unique’ file that is created, analogous to the original document handwritten by the author. Yet in the case of Hawking’s digital archive, it might be interesting for future historians and social scientists to study how Hawking’s digital files were reused in different contexts (Mialet, 2012, p 158), adapted for new uses, and circulated in his network and public interactions. This suggests a different emphasis for what is being studied, shifting attention from the generation of ideas towards their re-use, adaptation for new contexts and circulation. In observing digital culture, questions of replication, re-use, re-mix and dissemination do seem important with respect to digital archives. This is shown, for example, in Lisa Cianci’s (2017) study of digital artists’ archives, a shift also noted by Kirschenbaum (2016, p 192) regarding literary authors. In the case of the Hawking Archive, we therefore argue that the preservation and study of his digital files may facilitate research not just on the generation of his ideas and how he expressed and communicated them but also how they circulated in his network and the wider world and how they were reused.

One possible reason why the Acceptance in Lieu offer did not include Hawking’s digital archive is the difficulty of valuing digital files. The easy replication and circulation of digital files, which might be considered one of their advantages for studying the circulation of ideas and texts in culture, are precisely reasons why they are hard to value, because they can be easily copied and thus cease to be rare or unique. This makes it difficult to include digital archives in acquisition processes that rely on valuation as in the case of the Acceptance in Lieu scheme. In commercial contexts digital assets such as digital artworks have been sold with the issue of a non-fungible token. This is a unique digital identifier that is recorded in a blockchain (a decentralised, and publicly accessible digital ledger) as a certificate of authenticity. It is not yet clear how effective this will be in holding the value of digital assets over time, although it is interesting to note that some archivists have also advocated blockchain as a potential technology to explore for retaining trustworthy decentralised and participatory archives (Findlay, 2017). In digital preservation practice a checksum – which according to the Digital Preservation Coalition (2015) is a unique numerical signature derived from a file and used to compare copies – is used to verify the identity of files over time. This, combined with a range of computer and human generated contextual information (dates, times, folder structure, file size, file format, creator), helps to preserve the authenticity and integrity of digital archives over time.

Another possible reason why the Acceptance in Lieu offer did not include Hawking’s digital files is concerns about unauthorised access to data. Many digital archives contain a large quantity of files and data, often combined with an apparent lack of structured organisation of files. This can lead to relevant, irrelevant, public, private and sensitive data becoming indiscriminately mixed in large volumes without human readable signposts to distinguish them. This can happen in large paper archives too, but as digital files are easier to replicate and circulate, the unknown sensitivity of data in large accumulations of digital files can be perceived to present more of a risk to unauthorised access. It appears from comments in the graduate assistants’ notebooks that the risk of unauthorised access was an explicit concern with respect to the Hawking digital archive. The examination of digital archives for data protected or sensitive data is a challenge that has been explored by archivists and digital preservation specialists including using different tools for automated searching to flag sensitive data as well as direct examination by archivists. Before providing access to digital archives, they are examined (including the identification of personal and sensitive information), appraised (and then retained or not retained), and catalogued, which may be an iterative process. Other processing may include file format identification and creating copies of files in currently accessible formats to view and access them. During this process, as is the case with non-digital archives, sensitive files can be reserved from access for specific time periods.

Technical challenges relating to the degradation of digital media also needs to be addressed. Data can be exactly copied from obsolete carriers onto current digital media. These files are reserved in storage as preservation master copies, without altering them so they can be referred to as the source file received in the archive. Storage of files in digital preservation systems is set up to be stable, replicated (in different locations) to provide backups and disaster-recovery, and is monitored and regularly migrated to current storage systems to prevent the degradation and loss of data. While digital archives continue to present many practical challenges to archivists and digital preservation specialists, approaches, standards and tools are available to guide the implementation of digital preservation systems and archives and there are many working examples in the UK. Perhaps the most important practical challenge is securing adequate resources to preserve and make digital archives efficiently and ethically accessible on a significant scale.

Despite the challenges, we think it is important to add digital files to the Stephen Hawking Archive, if possible, and we know that Cambridge University Library would welcome the opportunity to become the repository for this material. The scope and coverage of the Archive, especially from the period after the early 1990s, could potentially be extended. The potential for uncovering extra information preserved in digital files but not present in paper artefacts produced by printing out documents from digital files could be explored. The future study of research questions about scientific communication and the re-use, adaptation and dissemination of texts would be supported.

Conclusion

In this paper we have shown how the evolution of writing technologies and recordkeeping processes have shaped archival documents and the overall shape of the Hawking Archive. While Hawking’s specific circumstances led us to explore some of these themes, the Hawking Archive illustrates important aspects of working practices in groups within the scientific community and archival production. For example, the paper and digital tools required for mathematical physics, the collaborative nature of scientific work, the impact of technologies of communication on the nature and significance of archival documents as evidence of methods of working, the relationship between digital technologies and paper artefacts, as well as characteristics and challenges of the digital archive.

We hope this mediation of the Archive is not a closed process, and that processes of assembling, selecting and curating a Hawking digital archive may continue. In sketching out the concept, significance and possibility of a digital archive, we hope we have established its potential importance as well as our openness to reviewing some of the principles of appraisal that might accommodate it. For example, by paying attention to collaboration, re-mix and dissemination in the archive, and reassessing what is valuable in this context. While handwritten notes and drafts in the Stephen Hawking Archive are evocative objects, this is not the only or even the main place to look for understanding Hawking’s methods of working, or the expressed traces of what was on his mind.

Acknowledgements

The Stephen Hawking Archive was accepted in lieu of Inheritance Tax by H M Government from the Estate of Stephen Hawking and allocated to Cambridge University Library in 2021.

Figure 1 is by Graham CopeKoga, reproduced by kind permission of Cambridge University Library. Figures 2–8 are by Mark Box and reproduced by kind permission of the Syndics of Cambridge University Library. Images 2 and 4–8 are reproduced by kind permission of the Hawking family. Image 3 is reproduced by kind permission of Sir Roger Penrose.

We thank the Hawking family, and colleagues at Arts Council England, Christie’s, the Department of Applied Mathematics and Theoretical Physics, Jonathan Wood and Cambridge University Libraries (with special thanks to Dr Jessica Gardner) for their contributions to the preservation of the Stephen Hawking Archive. We thank colleagues at the Science Museum Group for their cooperation throughout the Acceptance in Lieu process and the opportunity to submit this paper to the Science Museum Group Journal.