John Graunt, the law of decline and the origins of urban historical demography

29 November 2012   John Graunt, The Law of  Natural Decline and the Origins of  Urban Historical Demography   Professor Richard Smith   As the late Nobel Prize-winning British economist Sir Richard Stone put it in his magisterial set of Mattioli lectures, delivered in 1986, but published posthumously as Some British Empiricists in the Social Sciences 1650-1900 (Stone,1997), John Graunt was virtually a one-book man. But his short book Observations on the Bills of Mortality published in 1662 many would argue marked a sea change in the use of statistics and in the derivation of a methodology that was recognisably demographic. In that sense Johan Peter Sussmilch’s (Süssmilch, 1741) reference to him in 1741 as a ’Columbus’ in the role he performed seems very apt. When giving his pioneering lecture series in the 1920s and early 1930s, not a great distance from where we are meeting today at University College London, on the History of Statistics in the 17th and 18th centuries (but published only in 1978 under the editorship of Egon Pearson) Karl Pearson (Pearson, 1978) was in no doubt that John Graunt had a very real claim to be regarded as the father of statistics.   Be that as it may, the impact of the Observations was such that Graunt was very soon after its publication elected a Fellow of the Royal Society thereby bringing him within the famous Gresham College group. The Observations went into 5 editions by 1676 and quickly made an impact on his contemporaries in England such as William Petty and subsequently Edmund Halley as well as the likes of Huygens, Struyck and Leibniz on the continent to name just a few.   Graunt’s life history, his intellectual and personal relationships with William Petty and a disproof of the older argument that William Petty was the actual author of the Observations are issues that have been powerfully pursued by the late David Glass (Glass, 1963) in a classic paper actually prepared for a tercentenary conference on Graunt’s work at the Royal Society in 1962 (just 50 years ago). Interest in the work has been such that editions and annotated reprints have appeared again and again for instance by C. H Hull (Hull, 1899) in the late 1890s, F.W. Willcox (Willcox, 1939) in the 1930s and Peter Laslett in the early 1970s (Laslett, 1973).    Glass did more than most to describe and analyse the problems associated with Graunt’s tentative, but subsequently very influential attempts to distribute deaths according to age and thereby create the nearest thing to a life table. Ian Hacking (Hacking, 1975) has written elegantly in attempting to place Graunt’s work within a history of early ideas of Probability. Charles Webster (Webster, 1975), the historian of science and medicine, was able to place Graunt within that vast outpouring that flowed from the influence of Bacon’s Instauratio magna and The History of Life and Death. Linked to Webster’s work in his weighty Great Instauration are the reflections of Philip Kreager (Kreager, 1988) who has thrown a very bright light on Graunt’s methods by tracing them back to Bacon as well as relating his statistical techniques to common book keeping practices with which he would have been familiar as a London haberdasher and Kreager has gone some way further than most other historians of thought in explaining Graunt’s writings in terms of religious and mercantile ideas prevalent in the middle of the seventeenth century.   One other more recent commentator is worthy of mention, given that Graunt had much to say about plague in so far as it had a prominent place among the attributed causes of death reported in the Bills of Mortality and that is Professor Paul Slack (Slack, 1985). In his magisterial work on the impact of plague in Tudor and Stuart England Slack approaches the subject from the perspective of a historical demographer measuring the mortality impact and epidemic frequencies of plague as well as from that of a historian of thought who is interested in the way that ideas about plague came to influence public policy and in particular how plague which in the early seventeenth century had been viewed as having providential underpinnings came to be regarded as having distinctive attributes and predictable patterns as a result of Graunt’s application of Baconian dissection of the plague information in the Bills of Mortality and hence facilitating the acceptability of policies taken to thwart plague’s arrival and spread in England through the use of quarantine rather than a near exclusive reliance on prayer . As Slack notes Graunt made only one tentative reference to divine providence in the Observations and that was employed in order to discuss old ‘superstition’ about the timing of epidemics.   If time permitted I could add significantly to a long and distinguished literature on the Observations. One can certainly go so far as to claim that the template of statistical analysis of demographic data was provided by Graunt’s Observations. He offers us a way of engaging in critical analysis of the quality of data sets on which his calculations were based and establishes new standards in his consideration of mortality by cause of death as well as a demonstration of the stability of statistical ratios and through his efforts to construct a life table.   There is no time to go further into these matters which others have already considered most thoroughly. I want instead to focus on the issues raised by the urban context in which Graunt largely sets his analysis and from which he derives his data. Certain sections of the Observations retain significance of a high order in a set of on-going debates about the role played by urban centres in influencing geographically larger demographic systems and about whether such centres have inherent qualities that are common to all pre-industrial urban settings.   In Chapter VII entitled ‘Of the differences between burials and christenings’ Graunt offers the observation on the London Bills that there were far more burials than christenings … ‘for in 40 years from the year 1603, to the year 1644, exclusive of both years… there have been set down 363935 Burials, and but 330747 Christenings within the 97, 16 and 10 Out Parishes’ (Observations, p. 369)   ‘From this single Observation it will follow, That London should have decreased in its People; the contrary whereof we see by its daily increase of Buildings upon new Foundations and by the turning of great Palacious Houses into small Tenements. It is therefore certain, that London is supplied with people from out of the Country, whereby not only to supply the overplus differences of Burials, above mentioned, but likewise to increase its Inhabitants according to the said increase of housing.’ (Observations, p. 369)   ‘But if we consider what I have upon exact enquiry found true, viz. That in the Country, within 90 years, there have been 6339 Christenings, and but 5280 Burials, the increase of London will be salved without inferring the decrease of the People in the Country.’ (Observations, p.370)   The Country as defined by Graunt , in fact, is represented by the Bills for a parish in Hampshire, Romsey, in fact William Petty’s birthplace.   Graunt goes on to make various estimates of population sizes of London and the Country to support this relatively optimistic view of the capacity of non-metropolitan population growth to underwrite continued population growth notwithstanding the migration flows into London and the burials outside of England of those emigrating to the Colonies or lost abroad as fatalities in military conflicts.   Graunt probes further into the differences he had detected between city and country and attempts to show ‘why although in the Country the Christenings exceed the Burials. Yet in London they do not. The general Reason for this must be that in London the proportion of those subject to die, unto those capable of breeding is greater than in the country… let there be an hundred Persons in London, and as many in the Country; we say, that if there be sixty of them Breeders in London, there are more than sixty in the Country or else we must say that London is more unhealthful, or that it inclines Men and Women more to Barrenness than in the Country’ (Observations, p. 372)   He sets out why there were proportionately fewer breeders in London than in the country since London received many men coming there for reasons of justice and trade who left their wives at home in the country as did persons coming to London in search of pleasure or cures for ill health and that many males who were bound as apprentices did so for 7-9 years when marriage was unavailable to them. Finally there were many sea men in London who left there wives unable to breed without them which might have encouraged promiscuity.   Graunt does draw a distinction between the relative numbers at risk to breed in town and country and the unhealthiness of the two settings. He notes that ‘seasoned bodies may and do live near as long in London as elsewhere, yet new-comers and children do not for the Smokes, stinks and close Air, are less healthful than that of the Country’ (Observations, p. 373). This observation he feels is justified by what he sees as the tendency of sickly persons to retreat to the Country.   In reflecting on possible barrenness in London he suggests that intemperance in feeding, and especially Adulteries and Fornications are believed to be more frequent there and hindered breeding. Finally he feels that minds of men in London are more thoughtful and full of business and thereby subject to anxieties whereas in the Country corporal labour and exercise promote breeding    Being very alert to the quality of the data with which he worked —a feature that was a hallmark of his approach--he did spend some time in reflecting on the accuracy of the Christening record available to him. He ponders the ‘neglect of Christenings’ and believed that the ‘accompt of Christenings hath been neglected more than that of Burials for reasons to do with religious opinion against the Baptizing of Infants that was particularly pronounced between 1650 and 1660, the scruples of ministers unwilling to baptise leading parents to take their offspring to those ministers who failed to or refused to register the baptism and the disincentive arising from the baptismal fee, although he does conclude that ‘Upon the whole matter it is most uncertain’. Furthermore he does not consider whether the ‘neglect of christenings’ was likely to have been greater in City than Countryside.   While the modern commentator may scoff at some of these attempts by Graunt to construct a causal account of contrasts between town and country we cannot avoid recognising that Graunt alighted in these parts of his Observations to which my attention has been drawn on a set of issues that form a core component of what is a central theme in present-day urban historical demography. To the extent that this issue was investigated at all, opinion could be seen as conforming to Graunt’s observations that there was indeed what Jan de Vries (1984) has called the law of urban natural decrease’. Johan Peter Süssmilch (Süssmilch, 1775) extended Graunt’s interests in these matters citing his evidence in the Observations but also alighting in part on calculations that Graunt’s near contemporary Gregory King had assembled in his mass of data to describe ‘The State of England in 1695’ (Thirsk and Cooper,1972) . King presented data showing that London’s crude death rate was slightly more than 41 per 1000, that of the lesser cities and market towns 33 per 1000 and villages lowest of all at 31 per 1000. The paradigm was retained with William Farr (1885) the great 19th century superintendent of statistics seeing this urban handicap arising directly as a function of viewing mortality levels as positively related to population density levels.    It is noteworthy that one of the twentieth century’s leading demographers Kingsley Davis Davis , 1973) subscribed to this view when wring in the 1970s and a quick perusal of the enormous data assemblage made by Roger Mols (Mols, 1954-6) in the 1950s in his formidable three volume Introduction à la démographie historique des villes d’Europe du quatorzième et dixhuitième siècle almost always shows towns exhibiting a burial surplus in the period between 1775 and 1850, although little data before that period on this matter are reported.   In fact there was a deeply embedded tendency to view city mortality as so high that whatever the fertility levels operating cities were unable to replace themselves without significant levels of immigration. The classic account for London was presented by Tony Wrigley in 1967 (Wrigley, 1967) when he demonstrated that because of London’s rapid growth against a background ( à la Graunt) in which burials significantly exceeded baptisms, migrations flows were such that perhaps by 1700 London required an annual inflow of 8,000 migrants to sustain the striking growth in its size that was occurring and that perhaps one sixth of all those surviving to adulthood in the whole of England would form this migration stream. The conceptual depth of the notion of natural decline in urban centres can be seen in Wrigley’s hugely influential book Population and History (Wrigley, 1969) that appeared in 1968 in which a mortality contour map is drawn showing an imaginary seaport as well as a smaller market town acting as demographic sinks for the larger rural hinterlands.   In 1978 the first real objection to the primacy of the ‘law of natural decline’ in the form that had been formulated in the time of Graunt came from Alan Sharlin (Sharlin, 1978) who was unwilling to accept that urban populations in early modern Europe would have inevitably declined without immigration. Sharlin to some extent turned the argument on its head by focusing on the migrants. He argued that city populations could be divided into two groups—permanent residents and temporary migrants. Among permanent residents births exceeded deaths while among the temporary migrants the opposite held. The permanent residents did achieve natural increase but the temporary migrants made up disproportionately of the young unmarried apprentices, servants and persons of marginal stratus were restricted in their capacity to marry and procreate, although they contributed in a significant manner to the death totals recorded in city registers or Bills of Mortality. As de Vries (De Vries, 1984) puts it in Sharlin’s interpretation the frequently observed excess of deaths relative to the number of births occurred because of migration rather than being its cause. Sharlin used data from Frankfurt-am-Main to underpin his argument.    Another contributor to this debate was Ad van der Woude (van der Woude,1980) who was unwilling to accept the inevitability or universality of urban deaths as always exceeding births since he wished to argue that in most European cities by the early nineteenth century there were clear signs that births were beginning to exceed deaths, that is long before major campaigns for public health were bringing down urban death rates.   The difficulty confronting all of these positions was that the sophistication of the demographic measurements used in these debates was restricted in part because the tool of nominative linkage using family reconstitution that was pushing the analysis of demographic processes to higher and higher levels when applied to villages and small market towns were generally very hard to take to the highly mobile urban populations where accurate recording of vital events might also be compromised.   Nonetheless, use of the rudimentary methods that exploited demographic aggregates in the style of Graunt can reveal some startling results. For instance, in 1981 Wrigley now with his colleague Roger Schofield (Wrigley and Schofield, 1981) was in a position to place some firmer estimates around the London baptismal and burial totals and their relationship to those of England as a whole. From being just under 3 per cent of all baptisms those in London rose to marginally more than 12 per cent by the end of the seventeenth century. From just over 5 per cent London burials were up to 17 per cent of all those occurring in England in the early eighteenth century. In fact baptisms always exceeded burials in non-metropolitan areas but an overall national deficit was strongly associated with a period that overlapped with Graunt’s analysis with such large excesses of burials over baptisms in London that they took the combined London and non-metropolitan areas into a deficit situation. London’s role in bringing that outcome about could not be demonstrated more emphatically. De Vries (De Vries, 1984)comments with respect to these data that they are insufficient to fully reject Sharlin’s thesis since it is theoretically possible that the London age distribution and celibacy rates were sufficient to experience baptism totals well below burial totals thereby creating a mirage of excess mortality, although the assumptions would have to be extreme to generate so many deaths at ‘normal’ death rates—in particular a very youthful population within which deaths were heavily concentrated.   Data in this form also suggest that the circumstances of the later seventeenth and earl eighteenth when the burial surplus was particularly marked were less evident before 1650 and after 1750. Graunt himself was aware of changes across time in the extent of the excess of burials over christenings.   In chapter XII Graunt writes ‘When I consider, that in the country 70 are born for 58 Buried, and that before the year 1600 the like happened in London, I considered whether a City, as it becomes more populous doth not, for that very cause, become more unhealthful: am inclined to believe that London now is more unhealthful than heretofore, partly for that it is more populous, but chiefly because I have heard that sixty years ago few Sea-Coals were burnt in London which are now universally used’ (Observations pp. 393-394)   (John Evelyn’s Fumifugium with its plan for banishing ‘that hellish and dismal cloud of Sea-Coal’ was published in 1661.)   In raising the possible contrast between London at the end of Elizabeth I’s reign and the situation in the early years of the Restoration Graunt introduces another dimension to the discussion concerning the nature of London’s demographic regime and in particular the trends of mortality. In presenting the issues in admittedly over-simplified terms it is necessary to ask whether the notion of demographic sink and natural decline is effectively a constant with regard to large urban centres. In this respect some of the most penetrating recent work in historical demography as a whole has been directed to consideration of London’s mortality within the framework of the notion of an epidemiological regime.    John Landers (Landers, 1993) has been at the forefront of this work and from an empirical perspective in particular has done so by making use of the Bills of Mortality in the eighteenth century as well as a family reconstitution of London Quakers extending for two centuries after 1650. While he does not interrogate the same data sets that were central to Graunt’s work in the Observations he does bring into play a conceptual framework of analysis that helps us to understand the mortality patterns that were emerging and captured by Graunt as well those that were to follow in the subsequent century and a half after c1660. Landers was much influenced by ideas that were formulated initially by W.H McNeill, particularly a paper that he published in 1980 McNeill, 1980) in a relatively obscure collection developing more fully arguments seeded initially in his remarkably successful Plagues and Peoples (McNeill, 1977) These arguments give particular attention to the spatial structuring of society and their relative impacts on levels of exposure through such factors as population density and migrations flows over a range of geographical scales. Large metropolitan centres in the pre-industrial indeed pre-transitional (in a demographic sense) appear t especially amenable to interpretation within such a framework. In such settings densely distributed populations residing in poor housing, with poor sanitation and dependent upon polluted water supplies created what Landers has termed conditions where pathogens could be retained and conducted . In addition the location of such centres within dense networks of trade and migration treated a high exposure potential to such an extent that they functioned as endemic reservoirs of infection. As these centres grew mortality became correspondingly severer. But it did so, by impacting adversely on infants and very young children disproportionately. Those who survived the hazards of life in the youngest age groups entered later life with substantial immunological protection. Landers terms this configuration as a ‘high potential’ model of metropolitan mortality and while mortality in such a setting will be high it will not be distinguished by great volatility from year to year (unlike communities in the rural hinterland where mortality will be much lower but punctuated by severe epidemic outbreaks at irregular intervals). However those residing in the city will be distinguished by differing levels of immunological protection according to their age and their migratory history. Children whether native born or recent immigrants to the city will experience high mortality since they lack immunity to infections as will a nigh proportion of the recent migrants and in that respect will be lacking in immunities similar the very youngest. Adolescents and young adults who have been born in the city will also have acquired immunities as will those in the older age groups. The model does depend on a sizeable presence of ‘crowd diseases’ such as smallpox, measles and other respiratory infections that depend on person-to-person modes of transmission.   Landers provided some very persuasive evidence from his studies of the Quakers and use of evidence in the Bills of Mortality to show that in the early eighteenth century while infant and young child mortality attained an exceptionally high level (Infant Mortality Rates at 350-400 per 1000 (Landers, 1993, Laxton and Williams, 1989) compared with levels substantially below 200 per 100 in rural areas) while that of adults was falling which was a pattern that was also extended more broadly to the national level. In the metropolitan context an especially dramatic fall in infant mortality eventually occurred after 1750. It is thought to have been disproportionately connected with a substantial decline in first-month and neonatal mortality and charts a course paralleling that taken by maternal mortality which also takes a dramatic shift downwards after 1750. Migration assumes a major role in this account, although evidence bearing directly on the issue is not available to Landers. What he does, however, is to assume that most migrants to London were adolescents and young adults and gains confidence in that assumption because found that smallpox was the only disease reported in the London Bills strongly located as a cause of death among children and the age groups within which recent migrants were most likely to be found. Landers also uses the fact that among the Quakers those dying of smallpox over the age of 10 left no record of them having been born in London.    Since Landers published his ideas in the late 1980s and early 1990s piece meal work (eg. Newton and Smith, forthcoming) on infant mortality in both City and suburban parishes in London before 1700 has shown that there was clearly a net upward drift in infant and child mortality in both wealthy and poorer parochial locations which is consistent with the argument that mortality worsened in these age groups and did so strikingly in the period after the last plague outbreak in 1665. Landers shrewdly noted that it is perhaps easier to test the model in the period of worsening mortality than in that when improvements occur first to adults and then subsequently among the very youngest elements of the population. The Bills of Mortality reveal sharp falls in infant and maternal mortality in the late 18th century and this appears to be something that can be found across a reasonably broad social spectrum involving the general population, those in workhouses and those taken infants taken into the London Foundling Hospital (Levene, 2005) suggesting as Romola Davenport , Jeremy Boulton and Leonard Schwarz (Davernport, Boulton and Schwarz, 2011, unpub.a, unpub.b) have argued based on their recent and ongoing research on the large and rapidly growing parish of St Martin-in-the-Fields that these improvements were relatively egalitarian in their effects and that they are of an epidemiological type rather than specifically associated with a rise in the standard of living. Of course this does not rule out a role for smallpox vaccination as assisting significantly in this shift.   In certain respects in this argument as a whole migration, unlike in Graunt’s specification of the issue, cannot be regarded as some dependent variable in part responding to the shortfall of baptisms relative to burials, but to a certain extent becomes an independent variable with so much hinging on the age incidence and immunological status of those entering the metropolitan population from outside. In effect it is a mortality variant of the argument that Sharlin introduced over 30 years ago when he was privileging fertility attributes of migrants as the driving factor in determining the character of the demographic regime prevailing in large urban centres.    It could also be hypothesised, as have Davenport, Schwarz and Boulton (Davenprt, Boulton and Schwarz umpub.a) that there was a migrant ‘selectivity effect’ that was driving down the mortality risk with migrants to urban areas making up a selected population of healthy low mortality risk individuals. In a recently published paper in the Economic History Review in which they make use of the extremely valuable information in the sexton’s books of St Martin-in-the-Fields (Davenport, Schwarz and Boulton 2011) identify a sharp reduction in smallpox fatalities among adults after 1770, consistent with a sudden increase in exposure in rural areas brought about by a rise in the infectiousness of the smallpox virus which made those migrating from such areas into London arriving with far lower risks to their life chances.   These issues are still thinly researched in part because of the relative difficulty of finding demographic data that enables the simultaneous study of mortality changes by age along with evidence bearing upon the cause of death. However these changes seem clearly implicated in a major, indeed one might claim revolutionary shift in the epidemiological environment in London and other urban centres since there is evidence that not just in London but other urban centres in England and elsewhere in north-west Europe an urban demographic transition gets under way after 1770 with urban mortality rates falling below urban fertility rates. Jan de Vries noted that until this is established on a permanent basis urban growth depends entirely on migration. It is a crossing point that must be passed to remove a ceiling on overall urban growth . De Vries (1984) reminds us that until that point is reached as simulation exercises undertaken by Nathan Keyfitz (Keyfitz and Philipov, 1981) and others show the urban sector is unable to surpass a maximum of 40 per cent of the total population imposing severe constraints on sustained economic growth.    It can be claimed that understanding the epidemiological underpinning of the shift from the world that was described so precociously by John Graunt in his Observations upon the Bills of Mortality to the one that is being unearthed in the more recent work on the mortality regimes of large urban centres constitutes one of the most pressing tasks requiring the attention of historical demographers or those who might be more appropriately described as historical epidemiologists.   References: Davenport, R., Schwarz L., Boulton J., ‘The ecline of aduklt smallpox in eighteenth-century London’, Economic History Review 64 (2011), pp. 1266-1288 Davenport, R., Boulton J., and Schwarz L., ‘Infant and young child mortality in Lonson;s west End, 1750-1824’ (Unpub.a) Davenpoort, R., Boulton J., Schwarz L., and Black J.,’ Neonatal and maternal mortality ion the workhouse of st Martin-inthr-Fields 1725-1824’ (Unpub.b) Davis, K. ‘Cities and Mortality’ International Population and Urban Research, Institute of Internal Studies, University of California, Berkeley, reprint no. 433 (1973) De Vries, J. European Urbanization 1500-1800 (1984) De Vries, J., ‘Problems in the Measurement, Description, and Analysis of Historical Urbanization’. In A. van der Woude, J. de Vries and A. Hayami (eds) Urbanization in History: A Process of Dynamic Interaction (1990) Farr, W., Vital Statistics (1885) Glass, D.V., John Graunt and his natural and political observations’, Proceedings of the Royal Society B 159 (1963), pp. 2-32. Hacking, I, The Emergence of Probability (1975) Hull, C.H., (ed) The Economic Writings of Sir W. Petty vol II (1899)  Keyfitz, N., and Philipov, D., ‘Migration and Natural Increase in the Growth of Cities’, Geographical Analysis 13 (1981) Kreager, P., ‘New Light on Graunt’, Population Studies 42 (1988), pp. 129-140 Landers. J. Death in the metropolis: Studies in the demographic history of London 1670-1830 (1993) Laslett , P. (ed) John Graunt’s Natural and Political Observations made upon the Bills of Mortality (1973) Laxton, P. and Williams N., ‘Urbanization and infant mortality in England: a long term perspective and review’, in M.. Nelson and J. Rogers (eds) Urbanisation and the Epidemiologic Transition (1989) Newton, G. and Smith, R., ‘Convergence or divergence? Mortality in London, its suburbs and its hinterlands between 1550 and 1700’ (forthcoming, Annales de Démographie Historique) McNeill, W.H. Plagues and Peoples (1977) McNeill, W.H., ‘Migration patterns and infections in traditional societies’, in N.F. Stanley and R.A. Joske (eds) Changing Disease Patterns and Human Infections (1980) Mols, R., ‘Introduction à la démographie des villes d’Europe du 14e au 18e siècle, 3 vols. (1954-6) Pearson, K., The History of Statistics in the 17th and 18th Centuries (ed. E. Pearson, 1978 Sharlin A., ‘Natural decrease in early modern cities: a reconsideration’, Past and Present 79 (1978) Slack, P. The Impact of Plague in Tudor and Stuart England (1985) Stone, R. Some British Empiricists in the Social Sciences 1650-1900 (1997) Süssmilch, J, P. Die göttliche Ordnung in den Veränderungen des menschlichen Geshlechts aus der Geburt, dem Tode und der Fortpflanzung desselben erweisen (1741) Süssmilch, J.P. Die gõttliche Ordnung in den Veränderungen des Menschlichen ( 4th ed.,1775)  Thirsk, J and J.P. Cooper (eds) 17th Century Economic Documents (1972 Van der Woude, A.M., ‘Demografische ontwikkeling van de noordelikje Nederlanden 1500-1800’, Algemene geshiedenis der Nederlanden vol V (1980) Webster, C., The Great Instauration: Science, Medicine and Reform 1626-1660) (1975) Willcox, F.W. (ed), John Graunt’s Natural and Political Observations made upon the Bills of Mortality (1939) Wrigley, E.A., ‘A simple model of London’s importance in changing English society and economy’, Past and Present 37 (1967), pp. 44-70 Wrigley, E.A., Population and History (1969) Wrigley, E.A. and Schofield, R.S. (1981), The Population History of England 1541-1871: A reconstruction (1981)       © Professor Richard Smith 2012