The Identity Project: an Assessment of the UK Identity Cards Bill and its Implications, produced at the London School of Economics (LSE), is one of the most exhaustive analyses of a contemporary policy proposal carried out by social scientists in recent decades. Exhaustive - except from one perspective: history. British identity card systems have been introduced twice, between 1915-1919 and 1939-1952. While the authors of the LSE report were aware of the Second World War card, they made no reference to the earlier card, and no attempt was made to learn substantially from either experience. Partly, no doubt, the vast technological gap between previous cards and contemporary proposals is one explanation for this lapse. The historical card systems were very simple: a folded card carrying a deliberately limited set of information backed up by registers of further personal and administrative data, held locally or centrally, and processed by hand. The contemporary proposal is for a highly sophisticated smart card, with biometric and other personal data embedded, supported by a computerised central register.
Nevertheless, the administrative operation of - and public response to - the historical card systems reveal features that should make all parties in the contemporary debate pause for thought. For example, the relative technological simplicity of the old card systems made a considerable contribution to their effectiveness: the simplistic equation of technological sophistication with effectiveness should be resisted. This paper briefly describes the introduction, operation and fate of the two historical British identity card systems, before turning to identify the features that have strong current political relevance, and drawing out their policy implications.
Before the First World War, Beatrice Webb could list, with some exasperation, the existence of eighteen different, disjointed registers of personal information in use by governmental agencies. Partial registers of personal information - systems like local electoral registers, lists of national insurance contributors, and so on - were indeed typical of administrative information systems in Britain. Webb, like many supporters of the contemporary identity card, assumed that such partial registers were inefficient compared to the potential of a single, universal register.
The first National Registration was taken during the First World War. The context was a fierce debate raging in the War Cabinet between those ministers willing to consider conscription and those who wanted to continue the policy of voluntarism. The argument turned on knowing the number of men within the population available to fight, and existing statistics were judged to be insufficiently accurate. Remarkably, given that this was essentially a census-type question, the Cabinet decided to resolve the matter through the introduction of national registration. Under the National Registration Bill, introduced by the President of the Local Government Board, Walter Long, in July 1915, personal information on all the adult population was compiled in locally-held registers, and identity cards were issued. The figure that interested the War Cabinet was soon generated: 1,413,900 men in England and Wales were still available for national service. Once this figure was found, politicians' interest in National Registration, and therefore also the identity card, dramatically waned. Cards were lost, left in pockets and the backs of chests of drawers. By July 1919, the Manchester Guardian could remark:
Apparently, National Registration had survived merely in order to be forgotten, and the news that the Government no longer desires that it should be kept up to date sounds rather like reading the funeral service over a mummy. [Yet] the National Register was a big enough Pharaoh in its day...its taking went far to overshadow more combatant aspects of the war, and August 15th, 1915, was as big a day for contemporary vision as September 25th and the opening of the luckless Loos offensive.
British civil servants did not forget the identity card. Plans for a new national register were written into the War Book. On the 29th September 1939, the second National Register was introduced for three specified purposes 'for the duration of the present emergency': co-ordinating national service, national security and the administration of rationing. This time the local registers were backed up by a comprehensive central register held at the Central National Register Office near Southport. Seven thousand transcript books contained details of forty million registrations. Identity cards - folded cards with name, address but not date of birth - were issued.
Crucial to the operation of the second National Register was its intimate connection to the organisation of food rationing. In order to renew a ration book, an identity card would have to be produced for inspection at a local office at regular intervals. Those without an identity card, would within a short period of time no longer be able, legally, to claim rationed food. This intimate connection between two immense administrative systems was vital to the success of the second card - they were not forgotten by members of the public - and provides one of the main historical lessons, discussed below. However, this connection also had consequences for the criminal uses of identity cards in Britain.
By the early 1950s, the identity card had become a routine part of policing. In 1950, a young man, Clarence Willcock, was stopped in his car by a policeman in North London on suspicion of speeding and asked for his identity card. Like the good Young Liberal that he was, Willcock refused to produce his card. Willcock's argument to the Middlesex magistrates was that the National Register was a piece of wartime legislation that was no longer in force in peacetime. The magistrates disagreed. The Appeal Court not only confirmed the judgement but also gave Willcock an absolute discharge and in his concluding remarks, Lord Goddard, the Lord Chief Justice and soon to be infamous as the hanging judge in Derek Bentley case, strongly criticised the police's use of identity cards:
It is obvious that the police now, as a matter of routine, demand the production of national registration identity cards whenever they stop or interrogate a motorist for whatever cause. Of course, if they are looking for a stolen car or have reason to believe that a particular motorist is engaged in committing a crime, that is one thing, but to demand a national registration identity card from all and sundry ... , for instance, from a lady who may leave her car outside a shop longer than she should, or some trivial matter of that sort,...is wholly unreasonable.
This Act was passed for security purposes, and not for the purposes for which, apparently, it is now sought to be used. To use Acts of Parliament, passed for particular purposes during war, in times when the war is past, except that technically a state of war exists, tends to turn law-abiding subjects into lawbreakers, which is a most undesirable state of affairs. Further, in this country we have always prided ourselves on the good feeling that exists between the police and the public and such action tends to make the people resentful of the acts of the police and inclines them to obstruct the police instead of to assist them.
To paraphrase, facetiously: 'Don't let 'em have it'. Such criticism, when combined with the gradual removal of rationing restrictions, made the National Register unsustainable. However, it should be noted that the end of rationing alone would probably not have ended the second British identity card. At that moment the National Register was being used to generate the National Health Service Central Register and it is more than plausible that a second intimate connection between administrative systems would have replaced the former one between National Register and food rationing.
There are many lessons to learn from the two experiences of identity cards in British politics. I have grouped these lessons under seven headings: effectiveness, moral panics, value for money, 'function creep', technology, national identity, and crime.
The first national register was a dismal failure, while the second a partial success. The designers of the second system learned from history. Sylvanus Percival Vivian had watched from the sidelines the fate of the first card, and, as Registrar-General had the responsibility to design the second system. In 1934, Vivian pinpointed the essential feature - 'parasitic vitality' - of a successful identity card:
if it cannot be given enough real peace value of its own it must be given a borrowed and artificial peace value...its use and production and the quoting or recording of the number upon it must be made obligatory in regard to as many as possible of the organised activities in close touch with the life of the people. If it has not sufficient vitality of its own, it must derive a parasitic vitality from established national institutions and social organisations...
Any system of [National Registration], as being an instrument of conscription, would obviously be received by the public with some reserve and suspicion, and in its actual administrative working, when established, would be exposed to a hostile bias on the part of the individual members of the public. By linking that system with the equally necessary system of registration for food rationing purposes...motives would be interlocked.
The purpose of the second identity card system was clearly understood and articulated. Once the political backing was secure, civil servants such as Vivian identified weaknesses - in particular the public's reluctance to keep the card - and designed solutions to overcome them. In contrast, the purpose of the first card had been poorly understood and articulated. It was driven by a Cabinet argument that, once resolved, left the card high and dry. Contemporary proposals for identity cards bear much more similarity in this respect to the first card than the second. Again, we find the purpose of the card remarkably hard to pin-down. Certainly, there is no evidence that measures to give the contemporary card a 'parasitic vitality' are being given the thorough consideration necessary for success.
One lesson to be drawn from historical experience is that universal registers of personal information are held to be solutions to moral panics, but in operation they are very rarely as effective as their proponents hope. Partial registers, such as registers of licensed car drivers or national insurance contributors, have not generated anything like the exaggerated hopes that universal registers have. For example, during the Second World War many commentators held that identity cards would stop bigamy. In fact, bigamy was neither particularly prevalent nor was it prevented by the existence of a national register.
A quick list shows that there has been no settled view of the targets for the contemporary card schemes. New Labour has proposed identity cards first as tools for combating underage drinking, then, as a tool against identity theft, illegal working and benefit fraud (the 'entitlement card'), and, more recently, as a tool against terrorism. However, all parties in the current ID card debate should be aware that in the periods outside 1915-1919 and 1939-1952 the potential of identity cards has been latent, and it has been all too easy to overestimate the size of this potential.
The threat posed by the Willcock vs. Muckle case to the second identity card prompted civil servants to examine in great detail how far the National Register offered value for money. This exercise gives us an in-depth understanding of the full value, and uses, to which the National Register was put. The net cost of the National Register was estimated at 800,000 per annum (800,000 in 1950 was equivalent to £17,841,000 in 2004, using the RPI). 'Supposing National Registration did not exist', the civil servants asked, 'what would be the cost of maintaining such of its functions as could be performed otherwise?'. Irreplaceable functions chiefly consisted of the ability to 'report the address of, or send a letter to, any person whom the authorities desire to find' including: criminals, escaped mental patients, reservists called up in time of war and so on, including one censored function. In total the measurable cost of discontinuing the National Register was estimated at 222,000, while the value of services that only the Register could provide was thought to be between £300,000 and £500,000. The remarkable conclusion is that even when calculated by friendly civil servants, the National Register was not value for money: it cost £800,000 and saved between £522,000 and £722,000. Aside from security service uses, which cannot be adequately assessed, the critical argument that saved it was that, in the event of war, the Register would cost another million pounds to reintroduce - and, during the Cold War, that was not unlikely.
A much simpler National Register and identity card system was barely, if at all, value for money. Many of the applications of this system are identical to those imagined under contemporary proposals, which would be, of course, vastly more expensive. The LSE group estimated the cost of current proposals to lie between £10.6 billion and £19.2 billion, with a median of £14.5 billion, assuming that the scheme would take ten years to implement.
The 1939 Act provided for three administrative applications of the personal data held in the registers (national service, national security and food rationing), but eleven years later thirty-nine government agencies made use of the records. Some of these uses were noted in the last section. Others involved the opening of Post Office Savings Bank accounts, collecting parcels, checking pension claims and routine police inquiries. Theoretically, exceptional disclosure was confined to communication of information relating to serious crime or national security. In practice, requests from government departments and state agencies relating to less serious matters were often granted. Disclosure was powerfully shaped by the culture of discretion that marks the British civil service, and was, of course, entirely free of constraint by data protection laws.
In general, the second identity card system illustrates the phenomenon of 'function creep', where seemingly insignificant further uses are incrementally added until, eventually, the pattern of disclosure and use of personal information is markedly different from that originally declared. Once a universal register was in place, there was a seemingly compelling bureaucratic case for using the information.
The most profound difference between historical identity cards and contemporary proposals is technological. A smart card holding biometric data (iris scans, fingerprints and so on) seems a world away from folded stiff paper. However, the simplicity of the second identity card was crucial to its operation. To be precise: it was the contrast between the limited amount of information held on a personal card and the greater amount held on the registers that permitted the operation of the 'checking' system, the aspect of the second national register that its architect, Vivian, was most proud of. When presented with a dubious card an official could interrogate the bearer, asking for a date of birth. Since the date of birth was not on the card, the bearer would have either to guess or have somehow found out the correct information. In general, this check was a quick, efficient means of deducing whether the bearer of the card was indeed whom the card indicated. The advantage of the card's simplicity was such that Vivian resisted proposals to make the card more complex, for example by adding a photograph.
There are two features relevant to contemporary debate, relating to power and trust. First, as the operation of the second card amply demonstrates, the power of the system lies in the weight of distribution of information across the system: placing all the information on the card would leave the citizen in almost full control over its application, while keeping the card simple compared to the state's register shifts the balance of power away from the citizen. Second, a technologically-sophisticated card is more likely to be accepted on trust than a very simple card. But since the unfakeable card is unlikely ever to exist, this very trust is problematic: the people and organisations with the means and will to corrupt a biometric card are precisely the people and organisations it would be most dangerous to trust. Vivian's model, of simple cards and the checking system is a valid alternative.
Many continental European countries have had a long, continuous experience of identity cards. This history was the main reason why both British identity cards were debated in frames shaped by notions of national identity. Even the civil servants administering the first card system described it as a 'Prussianising' institution. Churchill, in a Commons debate on 3rd September 1939, regretted the trespass made on 'our dearly valued traditional liberties', and looked 'forward to the day, when our liberties and rights will be restored to us, and when we shall be able to share them with the peoples to whom such blessings are unknown'. Public commentators were rarely shy in pointing out how the card conflicted with ideas of what in meant to be British.
While debates over contemporary proposals are often introduced with the observation that identity cards exist in many countries, they are not structured, to anything like the degree found in the historical cases, by a sustained opposition of 'British' and 'un-British'. The reasons for this shift are as yet undetermined. One factor may well be the cosmopolitanisation of British society, not least increased travel to the continent building familiarity with card-carrying cultures.
One final difference between the second card and contemporary proposals is the recent assumption that the broad nature of political society is not going to change radically. This assumption could not have been made in 1940, say. Indeed, a very rare direct reflection of this state of affairs can found in the value for money review. In weighing up the pros and cons of the National Register, the (unnamed) civil servant considered what he or she called the 'totalitarian" argument': 'while it is true that if this country went communist or fascist the National Register would prove a very handy means of finding any individual whom the authorities did not like, its uses in throwing up classes of individuals would be much more limited. The National Register cannot pick out the Jews or the bourgeoisie or the Roman Catholic priests or the agents or members of any political party. All it can pick out is aliens, without distinction of nationality, and persons of a particular sex and age, and persons who had a particular occupation'. I think it is fair to say that Home Office discussions about the identity card do not begin 'if this country went communist or fascist...'
The Daily Express was entirely correct to note in March 1945 the creation of a 'new underworld industry of faking, stealing and selling identity cards'. Recent research has uncovered a number of specific cases, which illustrate the diverse and often unpredictable unofficial uses of identity cards. The cards were relatively easy to fake One case, for example,
revealed an ingenious if rough method of forging a rubber stamp. Concentric circles are made in indelible pencil with a penny and a sixpence. In the space between the circumference the desired wording is inserted backwards. The paper is then wetted, and pressed hard on the official document. If the letters are written by a good draughtsman the impression left is surprisingly effective
A typical fraudulent use of the identity card was the case of 'T':
T obtained possession of X's Identity Card and used it to travel the country giving non-existent addresses or addresses of bombed-out buildings. . . At nearly every district he visited he perpetrated some fraud either on public assistance or on some private person. Our own procedure proved useless to trace the man, since he had always moved on before the police for his last area of operations could be informed, and the address was in any event fictitious. He was finally caught by the operation of the Assistance Board black list when he appeared at their Fulham office, and was sentenced to two years' imprisonment.
But there were also many unusual, unexpected cases that demonstrate that cards could become involved in complex constructions of personal identity. For example:
R. applied for a 5th issue Ration Book and fortnight later applied again as F. producing a second Identity Card issued to F. on the same schedule. An alert clerk recognised him on the second occasion. When interviewed by the police R. gave the most circumstantial account imaginable of F.'s identity, activities, friends, business, hobbies and tastes, ... and a plausible explanation of why he held F's Identity Card on his behalf. It was only after the lengthiest cross-examination conducted by the interviewing constable with high forensic ability that it emerged that F was fictitious and R had entered himself twice on the same schedule under different names. R emerges either as schizophrenic or a liar ranking in intelligence and noble inventiveness with Munchausen. Moreover, he had the last word. To the policeman's culminating question 'Does then such a man exist as F.?' R replied 'To me he does'
Cards were used to create proof of extra children, to claim extra ration books, to allow deserters to pass undetected, to make fraudulent assistance claims, and to enable people to serve in the armed services underage. In the first half of 1945, there were 217 prosecutions under the National Registration Act for false declaration, 398 for impersonation, 113 acts of falsification or forgery, 61 for allowing others to use another's card and 433 thefts. These figures, of course, only record cases of successful prosecution and just give an idea of the bare minimum of identity card criminality. There was only one case of a spy being (nearly) caught through the card checks - and Vivian 'in the course of his contacts with officers of the Security Service...reached the personal conclusion that the negative results were due to the absence of enemy agents', not the failure of the system.
It is difficult at the stage to estimate the extent that the proposed new identity card would be put to criminal uses. The main lesson from history, perhaps, is that the past perpetrators were highly inventive in finding unofficial uses, and that many of these uses were unanticipated by officialdom.
This paper has examined the two experiences of identity cards in British history and identified features relevant to contemporary debate. Public indifference or hostility to the identity card was managed by building 'parasitic vitality' into the system. In particular, the system of national registration was intimately connected to the system of food rationing. Without similar 'parasitic vitality' contemporary proposals can be expected to struggle to win acceptance. However, such interconnection encourages the phenomenon of 'function creep'.
Universal registration systems have repeatedly been proposed as solutions to short-lived moral panics. There is little evidence that national registers effectively resolve such panics. The last National Register, while relatively simple to operate and dependent on manual technology, was only marginally judged value for money when subjected to sympathetic but critical analysis. Part of the effectiveness of simple cards lay in their simplicity: the lack of information contained permitted a cheap and effective check system. Even so, the historical record also reveals the diverse unofficial, including criminal, uses of identity cards.
The first national register (1915-1919), and accompanying identity card, was a failure, and the second (1939-1952) a partial success. The success of the second system was secured by analysing the causes of the failure of the first. The latest proposals seem to resemble the first national register more than the second. Policymakers would do well to follow their predecessors and learn from the past.
LSE Department of Information Systems, The Identity Project: an Assessment of the UK Identity Cards Bill and its Implications, June 2005.
J. Agar, The Government Machine (Cambridge, MA: MIT Press, 2003).
J. Caplan and J. Torpey (eds.), Documenting Individual Identity: the Development of State Practices in the Modern World (Princeton: Princeton University Press, 2001).
LSE Department of Information Systems, The Identity Project: an Assessment of the UK Identity Cards Bill and its Implications, June 2005.
We are the only project in the UK providing access to an international network of more than 500 historians with a broad range of expertise. H&P offers a range of resources for historians, policy makers and journalists.