Alan Howkins & John Saville


The Nineteen Thirties:
A Revisionist History


From The Socialist Register 1979, pp.89-100.
Downloaded with thanks as a PDF file from the Socialist Register Website.
Transcribed & marked up by Einde O’Callaghan for the Marxists’ Internet Archive.

John Stevenson and Chris Cook published in 1977 The Slump: Society and Politics during the Depression: a volume whose purpose was to assess what the first chapter headlined as Myth and Reality: Britain in the 1930s. This is how the authors begin:

Of all periods in recent British history, the thirties have had the worst press. Although the decade can now only be remembered by the middle-aged and the elderly, it retains the all-pervasive image of the ‘wasted years’ and the ‘low dishonest decade’. Even for those who did not live through them, the 1930s are haunted by the spectres of mass unemployment, hunger marches, appeasement, and the rise of fascism at home and abroad.

These sentences preface a text which provides a markedly uneven treatment of the decade, in that it omits much that is relevant to its theme; which is altogether lacking in intellectual rigour; and which, for two young historians of the 1970s, is notably old-fashioned in its approach and techniques. The book, moreover, is a cobbling together of a good deal of material already published, by both authors. The social history chapters for example, which offer the central revisionist thesis of the volume, are no more than an expansion of John Stevenson’s chapter in Crisis and Controversy, Essays in Honour of A.J.P. Taylor (1976) edited by Alan Sked and Chris Cook. He uses the same quotations and the same sources; if new research has been undertaken, it is not evident. There would therefore be no point to a comment on this present volume were it not for its uncritical reception by some of its reviewers, to whom certain of its elementary points appeared to come as a blinding light of revelation; and also, because the book does raise some questions that need to be taken further.

What apparently came as a surprise to some reviewers was the discovery that the thirties were not a decade of unrelieved gloom, but that for those who remained in work real wages were higher than ever before. These facts, it must be said at once, have been the stock in trade for many years for both economists and historians. C.L. Mowat, in 1955, in Britain between the Wars, was fully aware of the position and Sidney Pollard in The Development of the British Economy 1914-1950 (1962) provided detailed statistical information. And all this before Aldcroft and Richardson began writing. Popular belief, including the understanding of many within the British labour movement, has accepted the mythology of crisis and unrelieved depression; and for these misconceptions socialist historians are no doubt especially to blame. The uneven, but unmistakable upward trend of working-class living standards, during the inter-war years were part of the secular improvement in the material basis of ordinary life characteristic of all mature industrial societies in the twentieth century since 1914; but one must always accept the qualification of unevenness as between different social strata and different social groups within the working class. The causes are well known: among them, the rise in output and productivity; technological change and occupational differentiation, particularly the shift away from wholly unskilled jobs in the manufacturing sector; the emergence of a large tertiary sector – with a large unskilled component; the increase in the propensity to consume; the decline in family size. How much improvement there was in different countries down to the beginning of the Second World War partly depended on the poverty levels and national income per capita in earlier periods, partly upon the course of economic change in the inter-war years and not least upon the incidence of world crisis and depression at the end of the 1920s. For Britain between the wars there were factors on both sides of the equation. After 1920 the economy suffered from an ‘over commitment’ to certain staple export trades in the nineteenth century, with the processes of adaptation to a new world situation being long drawn out and very painful in terms of human costs; hence the depressed areas with their permanently higher than average levels of unemployment. The 1920s saw the economy labouring under the additional burden of an over-valued currency; but after the low point of the slump in 1932 the combination of a housing boom and the growth of the ‘new’ industries (electrical engineering, motor vehicles, food products and a range of consumer goods marketed through the growing distributive trades) led to a considerable expansion of the economy with its high point in 1937 – in which year official unemployment was still above the million mark. There had also taken place a significant and favourable change in the age structure of the population, with a now larger proportion in the producing age groups of fifteen to sixty-four. But it was the dramatic fall in food costs which contributed most to the improvement in real wages in the thirties, given the high expenditure on food in relation to total income among all working-class groups compared with forty years later.

To underline the improvement in real wages and real standards of living over the 1913 levels requires, however, important qualifications if we consider the working population as a whole. There are different measures of this improvement – between ten and fifteen per cent in the 1930s for those who remained throughout in full-time work; but it is difficult to be precise about exactly how many did in fact remain continuously employed. If for example the earnings of men in work were spread over the whole of the working population – employed and unemployed – the position in the thirties would compare far less favourably with that of pre-war: the improvement by 1937 would only be around five per cent. But the most significant factor – and one which it is very difficult to be at all precise about – is the much greater seasonality of employment in the inter-war years. Set aside for the moment the matter of the cyclical fluctuations in demand for labour, and their consequences; and recognize that seasonal fluctuations, in the best years, have always been built into the labour markets of industrial capitalism until the post-1945 period. The building industry has always known seasonality, as did motor car workers in the years between the wars. The statistical calculation of the impact of seasonality upon real standards of living is rarely if ever made. It has been largely ignored, for example, in the Hobsbawm-Hartwell standard of living debate for the first half of the nineteenth century, among other reasons because of the absence of much of the relevant data. What kind of guesses can be made, for instance, about the largest working group before 1850, the agricultural labourers, divided between fully employed workers throughout the year and those on a casual basis, which could be daily or weekly or sometimes longer but short of an annual contract? Similarly for the 1930s. Casual, semi-casual, seasonal work entirely alters the calculations of real wages and real standards. Building workers, even in the best years, could nearly always expect at least the equivalent of a month’s loss of work through weather conditions or days lost in changing jobs or hold-ups of materials. And this in addition to almost all unskilled workers who are only rarely on permanent contract. As for those not subject to seasonality of employment in the 1930s, how do we calculate the improvement in real standards of living as against real wages? A comparison of the average real wage of a skilled engineer in 1913, 1929 and 1937 gives results which have been eagerly seized upon by conservative historians; and of the improvement over time there is no doubt; but by how much? Real wages and real standards of living are different matters. If a skilled engineer was unemployed for three years between 1930-1933, and then remained in work until the outbreak of war, had his standard of living risen by fifteen per cent by 1937-8? It is to be doubted. Jürgen Kuczynski’s work on labour conditions has either been rejected, because of ideological bias, or more commonly, ignored; but the questions he asked, and tried to answer in his statistical calculations remain important and significant; and there is here a major area of work for statisticians not crabbed and confined within conservative assumptions.

If we move from an assessment of material standards to the quality of life, the calculations become more complex. As E.P. Thompson wrote:

... the term ‘standard’ leads us from data amenable to statistical measurement (wages or articles of consumption) to those satisfactions which are sometimes described by statisticians as ‘imponderables’. From food we are led to homes, from homes to health, from health to family life, and thence to leisure, work-discipline, education and play, intensity of labour, and so on. From standard-of-life we pass to way-of-life. But the two are not the same. The first is a measurement of quantities: the second a description (and sometimes an evaluation) of qualities ...

It is quite possible for statistical averages and human experiences to run in opposite directions. A per capita increase in quantitative factors may take place at the same time as a great qualitative disturbance in people’s way of life, traditional relationships, and sanctions. People may consume more goods and become less happy or less free at the same time ... (The Making of the English Working Class, Penguin, 1968, pp.230-1)

A question of human experiences. Thus, to give one example from the 1930s, many of the car workers in the inter-war years – in one of the growth industries – were migrants from South Wales and other depressed areas. They left behind them family relationships and a way of life, a culture, which was not to be reproduced in their new environment. They also left behind unemployment; but like their forefathers in the 1830s, and the comparison is not far-fetched, they moved into a situation of social alienation. The Welsh migrated to Oxford and Slough: despised by the local inhabitants, they lived in social isolation in lodgings until they developed their ghetto communities. In work they moved into factories which were almost always non-unionized, often on piece rates and the conveyor belt. Arthur Exell was a young twenty-year-old Welshman who went to become a car worker in Oxford. His experiences have been published in History Workshop (Autumn 1978):

Mr. Kendrick was the top man for all the factory and he was the most hated man of all. He was terrible to work under, but we were frightened of him, and he was the boss of the factory. If you were one minute late in the morning he would send you back home ... I remember what happened once to Dickie Yates. He lived at Burford, which is about twenty miles away. And one day he got a puncture. He rode his bike all the way – even though it had a flat tyre – and there was Mr. Kenrick standing. He said, ‘Look Mr Kendrick, I had a puncture, I couldn’t do it.’ ‘Have another two days off to mend it’ said Mr. Kendrick. That’s the sort of pig of a man he was.

I remember one day I had completed my five radiators, just before the first hour was up. The inspector tapped the badge with his knuckles and shouted over to me, ‘Exell, all your cows have calved’, which meant the enamel of the cow had chipped out. This meant I had to take them all apart and rebuild them. The badges chipped very easily if you screwed them too tight, and would spin around if not tight enough. One man put cochineal in where the red had chipped; he was found out and got three day’s suspension. Because we were only allowed to work the number of hours specified for the job, I earned no money that day and when I got home to my digs I just sat down and cried. Things were very hard. (pp.53, 73)

Arthur Exell as a unit in calculations of statisticians belongs to the positive side of the standard of living equation in the 1930s, for all the material indices in his Oxford situation would show an improvement on those of an unemployed young worker in South Wales; but then, there is no good statistical measure by which to assess, in imaginative as well as rigorous fashion, a bullying foreman with seigneurial rights of hiring and firing; and in the absence of imagination and rigour, let us pass on the other side, holding firmly our pocket calculators, determined always to ignore all but the facts that Gradgrind himself would have recognized.

* * *

The old-fashioned character of the work under discussion is surprising. Women, for example, are subsumed under the discussion of men, and among many omissions Margery Spring Rice’s volume, one of the important sources for women’s condition, is not mentioned either in the text or the bibliography. Its absence fits the general approach. This is a book about an economic catastrophe which affected millions of working-class people. Their experiences have been written down in many places and in the past decade oral historians have begun to recover the tone, the feeling, the sentiment of life in the depression. But not in the Stevenson and Cook volume where the evidence is almost entirely based upon outside observers and where the poor are nowhere to be heard. Their voice is subjective; some things of what they are now saying about the past may be the product of forty years living with myth and illusion; but their testimony has a power and resonance which historians must assess and evaluate and which cannot be ignored, as it is ignored here.

It is not, however, only for its many lacunae that this volume is so unsatisfactory. The use of evidence is too selective, often dubious and sometimes wholly wrong and misleading. It is, in the final analysis, the authors’ scholarship which is in doubt. Here are a few examples. The first is taken from chapter five headed The Impact of Unemployment which on its second page (p.75) offers a critical contemporary quotation from a leading conservative historian, Arthur Bryant, on Orwell’s Road to Wigan Pier. Bryant dismissed Orwell’s account as propaganda. ‘Thus’, say our authors, ‘even for contemporaries, the picture was by no means clear. The true impact of unemployment upon social conditions was often obscured by the more emotive and committed writings on the subject.’ The contrary is the case. For the majority of middle-class contemporaries, certainly the majority of those who had access to the media – even more for those who owned the media – the position was all too clear: the unemployed were deemed to be work-shy, the dole was the scrounger’s charter, those who worried about the poor were soft-centred do-gooders. We hear these opinions again today, in the 1970s; but never in the volume or with the absolute certainty of the middle and upper classes of forty years ago. Historians who can write ‘even for contemporaries’ are not just naive: they have clearly failed to appreciate the range of opinion, to say nothing of the tone, of their chosen period. Nor is the reader helped by the slippery method of selective quotation. Dr M’Gonigle, for instance, was a well known medical authority who produced what became a famous report on the differential mortality between the employed and the unemployed in the early 1930s in Stockton-on-Tees; and the results were true for the standardized death rates as well as the crude. Our authors continue (p.79):

This result, however, was based upon a relatively small sample and not all commentators were prepared to accept its validity.

Now the reference to this sentence could be expected to reveal a source critical of the M’Gonigle results. It was in fact to G.D.H. and Margaret Cole’s The Condition of Britain, published in 1937. On p.95, under the heading Death Rates in Different Areas there is a discussion, among other work, of M’Gonigle’s researches, and after stating the results of the crude death rates analysis, they comment:

These results, though they are based on quite small investigations, are of real significance.

And after noting the revised figures based upon standardized death rates (that is, allowing for the age and sex composition of the employed and unemployed groups) the Coles add (p.98):

It is greatly to be regretted that many more investigations of this sort have not been made. The Stockton results are not put forward as yielding conclusions likely to be valid for the whole of the unemployed, for many unemployed persons have been out of work continuously only for short periods. Nevertheless what is true of Stockton may well be typical of a considerable section of those families whose breadwinners have been out of work continuously or nearly so for a stretch of years; and this question of the incidence of mortality among the families of the chronically unemployed especially deserves fuller study at the present time.

This is not quite what Stevenson and Cook told us, and the way in which they misleadingly summarize a careful argument is typical of their approach.

There are several political chapters, mostly of a psephological character, the work, we assume, of Chris Cook. In the first of these, Labour and the Working Class; the General Election of 1931, it is argued that the common belief of the catastrophic defeat of Labour being the result of the defection of MacDonald and his senior colleagues, is wrong. 1931 was not apparently a Red scare: the electoral defeat of the Labour Party was the inevitable consequence of first the general disillusionment of the voters with the record of the 1929-31 Labour Government, there having already been a series of by election defeats to underline the argument; and second, there was a desertion of the Liberal vote to the Conservative Party. There is some truth in both points, but typically of this volume, it is not the whole truth and indeed this sort of argument is fallacious in terms of an explanation of the landslide that occurred. Certainly the Labour Government had a miserable record, and the Labour Party would have lost any future general election, even without abnormal circumstances. But the situation was not normal from August 1931 until the polling day of 27 October. The break up of the Labour Government, the formation of a National Coalition and the appeal to the country at the end of October were parts of a political crisis that involved a pattern of misrepresentations, lies and abuse already well tried out in the previous coupon election of 1918 and the Zinoviev letter election of 1924. The Conservative vote in 1931 rose by over three millions; the Liberals’ fell by nearly three millions. The Labour Party dropped by two millions, their seats falling from 287 in 1929 to 46. It was a landslide victory for the National Government which had an enormous majority of 497 seats, with the Conservatives alone having 472 seats. But it was not, according to the authors, a panic election, and in three short paragraphs they exorcise the panic theory. (pp.101-2) Their last paragraph reads:

The whole assumption of a panic election has rightly been disputed. Bassett, for example, cites a comment from the Manchester Guardian of 29 October to the effect that ‘it has been a remarkably quiet election’ and though there were some reports of rowdyism these were very much isolated incidents.

It is typical that no page reference is given for this comment by Reg Bassett from his 1931: Political Crisis, published in 1958. It will be found on p.312 at the end of a short quotation from the Manchester Guardian which puts the ‘remarkably quiet election’ into a very different context from that suggested by Stevenson and Cook. Here are the three sentences from the Manchester Guardian quoted by Bassett:

There is agreement that it was panic, but the panic had not manifested itself beforehand in any open way. There had been no turbulent terror, none of the exasperation of temper in which panic often makes itself manifest. It has been a remarkably quiet election.

There was, in fact, a good deal of rowdiness on both sides, as Mowat noted; but elections in general became steadily less noisy and rowdy as the twentieth century moved along; and compared with most elections prior to 1914, that of 1931 was relatively quiet. But a landslide has to be explained with the shift of the Liberal vote on a massive scale to be analysed; and as Mowat again emphasised: ‘The most powerful weapon against the Labour Party was fear’ (p.410) to which Philip Snowden and MacDonald made their considerable contributions. The former’s famous speech ‘Bolshevism run mad’ – not mentioned by Stevenson and Cook – and the Post Office scare were important ingredients in the development of a desperate mood which swept the electorate into its remarkable support for the Conservative Party and the National Government.

* * *

One of the ways in which any political analysis of the 1930s must be judged is the approach and attitude towards the Communist Party, and more generally, towards the radicalization of sections of the British people during the decade. An adequate political sociology of British Communism has yet to be written, and whoever publishes it need not read this present volume for enlightenment. ‘The typical recruit of the mid-1930s tended to be young, middle-class and attracted by the appeal of the Left’ they write. But this is incorrect. It is true, of course, that one of the social components of the radicalization of the 1930s were university students, an important minority of whom joined the Communist Party. But to suggest that the radicalized student was the ‘typical’ recruit of the 1930s is to exhibit a wilful ignorance of social and political trends. The misconception is on a par with the common exaggeration of the part played by a coterie of radical poets around or associated with Auden; or the equally common failure to appreciate that the overwhelming majority of Britishers who joined the International Brigade were working class. John Cornford, Ralph Fox et al. were tragic deaths of enormously gifted individuals; but it was the solid working-class core of the Brigade, many experienced militants among them, who suffered the main losses. But to return for a moment to the ‘typical’ recruit of the 1930s: precise data are not yet available, but we do know the social groups from whom recruits must have come in relatively greater numbers than from the population at large. The miners of South Wales and Scotland, for example, and the metal workers of the Manchester region and London. Busmen in both London and the provinces; print workers in London and possibly elsewhere. And there are some defined groups besides the students who were radicalized. The Jewish community is an obvious example, and here we must include the Jewish middle class as well as the working class. There was always in Britain, going back to the 1880s, a radical trend among Jewish people in Britain; but the coming of fascism in Germany encouraged a much wider participation of Jews in the anti-fascist movement; and undoubtedly in areas such as the East End of London, or Leeds or Manchester, there was a significant accession of Jews to the Communist Party.

The Communist contribution to the political and intellectual history of the 1930s was considerable. Inevitably, Stevenson and Cook do not understand the problem, brought up as they have been in a very traditional style of historical research to which A.J.P. Taylor has contributed a great deal. The interconnections and interrelationships to be made demand an imaginative perception which Taylor, for one, deliberately eschews; and yet there can be no serious attempt at understanding without a willingness to interview, carefully and patiently, the historical record. It is, however, not just the cultural and intellectual record; there are many hard facts to be established of many different kinds.

One such fact is the attendance at political meetings. In his 1976 essay Stevenson discussed the significance of the Hunger Marches in the 1930s; and almost word for word the argument is repeated in this present volume under consideration. Briefly the argument is (a) the unemployed were only effectively organized by the National Unemployed Worker’s Movement, whose leader was the Communist Wal Hannington; and Communists formed the majority of the leadership throughout – agreed; (b) the unemployed in the 1930s were not ‘disposed to become revolutionary’ and (c) the great majority of demonstrations against unemployment were quite small affairs. The demonstrations and hunger marches organized by the NUWM ‘achieved very little and soon faded into obscurity’. (p.194)

These last two points demand a more detailed analysis than can be given here. We are dealing, not with intangibles, but with phenomena that are certainly difficult to be precise about. The testimony that contemporaries were impressed or influenced by the national hunger marches – the Jarrow March excluded – is not examined by Stevenson and Cook, not least because they never consider how public opinion is formed or altered. In the case of the hunger marches, it is necessary to begin with the reactions of the labour movement in its industrial and political sections; and to chart the remarkable change in sympathy and support for the NUWM’s marches between 1932 and 1934. In the latter year the local support for the marchers was considerable, and by 1936 this had reached the top levels of the movement, especially within the Labour Party. The London Trades Council officially backed the 1936 march, and at the demonstration in Hyde Park on 8 November 1936 a very large number of people gathered in Hyde Park to welcome the marchers. On the platforms, among Communist speakers, were Clem Attlee, Nye Bevan, Ellen Wilkinson and Edith Summerskill. It was raining most of the time, but everyone who was there thought it was a very large turn out; and so did most of the Press. Stevenson and Cook quote the police estimate at about 12,000, taken from the Metropolitan Police files in the Public Record Office. They continue: ‘It was an impressive gathering, at which the NUWM attracted far more support from the Labour Party than ever before’ (p.189), but on the following page, when pursuing their argument that support for the Hunger Marches was really quite small and much exaggerated at the time and by later commentators, they quote the police figure of 12,000 without qualification. This is what Stevenson had already done in his article of 1976.

There are several matters here that require comment. First, we are given no evidence from the Press or contemporary accounts to support the 12,000 figure. Second, and most important, why should police reports be accepted as good reliable evidence? Stevenson thinks they can be. In his 1976 article Stevenson wrote of these police statistics: ‘These figures appear reliable, coming from police records intended solely for internal circulation.’ (p.100) A somewhat naive statement for any historian to make, it might be thought. It is rather like believing the accuracy of Sir Nevile Henderson’s despatches from Berlin when he was ambassador to Nazi Germany. After all, his defeatist words were not intended for publication, but like the police statements, they were solely for internal consumption. This uncritical acceptance of police records and statistics, however, seems to be not uncommon these days; Skidelsky’s biography of Oswald Mosley for example quotes internal police memoranda at length (in the chapter headed The Campaign in East London). To have access to police records is of great value to the historian; but to accept them at face value, without check and counter check, is to abrogate the methods and techniques associated with serious historical research.

* * *

The debate on the 1930s in going to arouse much discussion, and socialist historians must not be absent from the controversies. What we need first of all is a much more detailed analysis of working-class living standards in all their complexities. We need as a first requirement to be able to quantify our results. How many? – is the central question that has to be answered when considering any statement about improvement or deterioration. There are immensely more data available than for the standard of living debate of the first half of the nineteenth century, and it is going to be possible to be fairly definite and precise about the material statistics of the argument. What socialist historians must not do is to allow the discussion to stop there. We are required to talk about attitudes, ways of life, social alienation and to present the whole man and the whole woman in the context of family life and class: and we must on no account allow ourselves to be bemused by those to whom the calculating machine and the computer offer substitutes for historical imagination.

Last updated on 4 July 2010