The road to the asylum: Institutions, distance and the administration of pauper lunacy in Devon, 1845-1914

Journal of Historical Geography

Volumn 25, Issue 3, 1999, Pages 298-332

  Melling, Joseph ^a,b   Turner, Robert ^a,b  

^a UNIVERSITY OF EXETER   (United Kingdom)

^b NONE

Author keywords

[No Author keywords available]

Indexed keywords

ADMINISTRATIVE SYSTEM; HISTORICAL GEOGRAPHY; INSTITUTIONAL FRAMEWORK; MENTAL HEALTH; NINETEENTH CENTURY; SOCIAL POLICY;

DEVON; UNITED KINGDOM;

EID: 0032821377     PISSN: 03057488     EISSN: None     Source Type: Journal    
DOI: 10.1006/jhge.1999.0117     Document Type: Article

References (100)

1. A. Scull, C. MacKenzie and N. Hervey, Masters of Bedlam: The Transformation of the MadDoctoring Trade (New Jersey 1996) 103-5 for an example of W. A. F. Browne's views on madness and modern civilization.
2. T. Harpham, Urbanization and mental health in developing countries: a research role for social scientists, public health professionals and social psychiatrists, Social Science Medicine 39 (1994) 233-45, particularly 240-1.
3. A. M. Clarke, A. D. B. Clarke and J. M. Berg (Eds.), Mental Deficiency: The Changing Outlook (London 1985). We owe this reference to Elaine Murphy.
4. J. K. Walton, Lunacy in the industrial revolution: a study of asylum admissions in Lancashire, 1848-50, Journal of Social History 13 (1979) 1-22,
5. and J. Walton, Poverty and lunacy: some thoughts on directions for future research, Bulletin of Society for the Social History of Medicine 38 (1986) 64-7.
6. W. H. Fisher and B. F. Phillips, Modeling the growth of long-stay populations in public mental hospitals, Social Science Medicine 30 (1990) 1341-4-7.
7. A. Scull, Psychiatrists and historical 'facts'. Part II: re-writing the history of asylumdom, History of Psychiatry 6 (1995) 387-94,
8. for a flavour of the exchanges; R. Porter, History of psychiatry in Britain, History of Psychiatry 2 (1991) 271-79, especially 270-74.
9. S. Craddock, Sewers and scapegoats: spatial metaphors of smallpox in nineteenth century San Francisco, Social Science Medicine 41 (1995) 957-68,
10. for discussion of Chinese community; E. T. Carlson, Nineteenth-century insanity and poverty, Bulletin of New York Academy of Medicine 48 (1972) 539-4-4, particularly 543, provides a useful comment on Jarvis's research on poverty, race and ethnicity in contemporary America from a positivistic perspective.
11. J. M. Hunter and G. W. Shannon, Jarvis revisited: distance decay in service areas of mid-19th century asylums, Professional Geographer 37 (1985) 299-300, especially Figures 1 and 2;
12. J. M. Hunter, Need and demand for mental health care: Massachusetts 1854, The Geographical Review 77 (1987) 139-56, particularly 153-54. Hunter provides a sophisticated analysis of a range of variables in these studies.
13. Social historians tend to emphasize the empirical limitations of Foucault's opus rather than its theoretical tensions but as Peter Dews has noted, there is no consistent analysis of institutions or their historical development in Foucault's work. Dews quoted in E. W. Said, Foucault and the Imagination of Power, in D. C. Hoy (Ed.), Foucault: A Critical Reader (Oxford 1986) 151.
14. F. Driver and G. Rose, Nature and Science: Essays in the History of Geographical Knowledge (Historical Geography Research Group, Research Series No. 28, 1992);
15. and D. Wishart, The selectivity of historical representation, Journal of Historical Geography 23 (1997) 111-118.
16. L. Guelke, The relations between geography and history reconsidered, History and Theory 36 (1997) 216-34,
17. provides a critical assessment on rationalist-idealist grounds in contrast to the Foucauldian critique of the 'natural history' tradition in empiricist geography. Some eminent historians have acknowledged the importance of space in historical development. See G. Blainey, The Tyranny of Distance: How Distance Shaped Australia's History (Melbourne 1966) for example.
18. C. Philo, 'Fit localities for an asylum': the historical geography of the nineteenth-century 'mad-business' in England as viewed through the pages of the Asylum Journal, Journal of Historical Geography 13 (1987) 398-415,
19. and Idem., Journey to the asylum: a medical-geographical idea in historical context, Journal of Historical Geography 21 (1995) 148-168.
20. See also H. Parr and C. Philo, 'A Forbidding Fortress of Locks, Bars and Padded Cells': The Locational History of Mental Health Care in Nottingham (Historical Geography Research Group, Research Series No. 32, 1996).
21. Philo, Journey, 150-1 and 154, citing G. N. Grob, Edward Jarvis and the Medical World of Nineteenth-Century America (Knoxville 1978), particularly 153.
22. Philo, Journey, 156-61. Philo also notes that there may be "some kind of urban-rural differential" in the pattern of admissions.
23. D. H. Alderman, Integrating space into a reactive theory of the asylum: evidence from post-Civil War Georgia, Health and Place 3 (1997) 111-22.
24. Alderman, Integrating space, 117 and passim.
25. Ibid., 119.
26. Alderman, Integrating space, 117 and 119.
27. The term 'local knowledge' is taken from C. Geertz, The Interpretation of Cultures (New York 1973). We use the term to suggest not merely the cultural boundaries of political discourse in particular regions (such as Ball) but also the micro-politics of societies such as Victorian England whereby specific forms of understanding and practices are associated with alternative and competitive views on a desirable social order. These rival 'traditions' included aristocratic responses to the modernization of the British state in provincial England and struggles around rationalities as diverse as models of political economy and psychiatry. For an excellent critical evaluation of Geertz's work,
28. see V. P. Pecora, The limits of local knowledge, in H. A. Veeser (Ed.), The New Historicism (London 1989) 243-76.
29. Philo, Journey, 162-3.
30. E. Jarvis, The influence of distance from and proximity to an insane hospital, on its use by any people, Boston Medical and Surgical Journal 42 (1850) 210-3 and 216-7.
31. Ibid., 216 and 220. Jarvis also discussed the Hanwell Asylum near London in his discussion of urban and rural institutions and their proximity to large cities.
32. Ibid., 219.
33. Philo, Journey, 158, 161-62 and 167 n52. Philo notes that the violations of 'Jarvis's Law' were explained by contemporaries in class terms and that the discourses developed in nineteenth-century England around the varied use of asylums were "spun around the axis of class". Also Philo, 'Fit localities', 3984-15, for a detailed discussion of Bucknill's contributions.
34. Philo, Journey, 157. Philo's suggestion that county asylums were usually placed at a point which was roughly central to the whole area, in contrast to American hospitals which were located in or near the largest towns and therefore remote from the frontier, must be open to question. Many such institutions were located within reach of the county town where certification by a magistrate was easily accomplished. Even the unusual Three Counties Asylum examined by Philo was close to Bedford. See
35. also F. Driver, Power and Pauperism (Cambridge 1993), chap. 1.
36. A. Scull, The Most Solitary of Afflictions: Madness and Society in Britain 1700-1900 (New Haven 1993) 12-38 and passim where Scull presents an ambitious model of the asylum's function within a society transformed by commercialization. Some differences emerged between Walton and Scull on the interpretation of class dominance and family strategies.
37. See Walton, Lunacy in the industrial revolution, 2-7.
38. The agency of families and lay opinion has gained considerable popularity amongst social historians of insanity in recent years. M. E. Keim, Women, families and the Provincial Hospital For The Insane, British Columbia 1905-1915, Journal of Family History 19 (1994) 177-193,
39. strongly emphasizes the therapeutic role of the family in contrast with the restrictive regime which derived from hereditarian principles in this province of Canada; P. E. Prestwich, Family strategies and medical power: 'voluntary' committal in a Parisian asylum, 1876-1914, Journal of Social History (1994), 802-4- and 806-7,
40. gives a less decidedly critical assessment of the asylum doctors' ambitions but also emphasizes the complex negotiations which engaged families particularly in relation to female admissions; A. Pernice, Family care and asylum psychiatry in the nineteenth century: the controversy in the Allgemeine Zeitschrift für Psychiatrie between 1844 and 1902, History of Psychiatry 6 (1995) 55-68.
41. For developed Weberian perspective on state power and territoriality, see M. Mann, Sources of Social Power I: A History of Power from the Beginning to AD 1760 (London 1986).
42. M. Poovey, Making a Social Body: British Cultural Formation 1830-64 (Chicago 1995) 73-131; and
43. J. Melling, The body in question: British cultural formation 1830-64, Journal of Victorian Culture 4 (1999) 164-73.
44. Alderman, Integrating Space, 115-16.
45. See A. Suzuki, Framing psychiatric subjectivity: doctor, patient and record-keeping in Bethlem in the nineteenth century, in J. Melling and B. Forsythe (Eds), Insanity, Institutions and Society 1800-1914: A Social History of Madness in Comparative Perspective (London 1999) 115-36.
46. Driver, Power, discusses a similar argument made by Giddens.
47. J. Melling, R. Adair and B. Forsythe, 'A proper lunatic for two years': pauper lunatic children in Victorian and Edwardian England-child admissions to the Devon County Asylum, 1845-1914, Journal of Social History 30 (1997) 371-405.
48. J. Melling, Accommodating Madness.
49. Carlson, Nineteenth-century insanity, 541-42,
50. discussing G. N. Grob's edition of Jarvis's Insanity and Idiocy in Massachusetts (Cambridge, Mass. 1971).
51. B. Forsythe, J. Melling and R. Adair, The politics of lunacy: the central state and the Devon County Lunatic Asylum, 1845-1914, in Melling and Forsythe, Insanity, Institutions and Society, 68-92, can be contrasted with Scull's influential studies of English insanity. Scull explains many of the peculiarities of institutional saturation and long-stay care in terms of the particular social dynamics of class society in England, though parallels can be drawn with American states.
52. In the case of Exeter, Plymouth and a small number of other historic boroughs, this responsibility for the Poor Law had been invested in a distinctive Corporation of the Poor since at least the eighteenth century, and they were therefore further insulated against the control of county magistrates and notables.
53. Throughout Devon as a whole, including the independent boroughs there were 200-300 pauper lunatics, usually referred to as 'idiots, imbeciles and lunatics' held within the various workhouse wards. The Poor Law unions of the administrative County of Devon held 130-140 pauper lunatics in the 1890s. See Annual Reports of the Lunacy Commission for returns on each county.
54. For reasons of consistency we have omitted those individuals whose parish of origin was not clear, including a very significant number whose residence was given as the workhouse of the Poor Law union and those transferred to Exminster from other asylums. The result was that some 2000 of our 13 000 cases were excluded. It is difficult to estimate the impact on our results but a detailed analysis of one Devon union indicates that residents of the workhouse were more likely to come from well-established families in areas reasonably close to the union workhouse rather than from transient groups of travelers or migrants. This raises a further set of questions about institutional dynamics that cannot be easily explored in the present paper.
55. See R. Adair, B. Forsythe and J. Melling, A danger to the public? Disposing of the pauper lunatic in Victorian and Edwardian England: the Exminster Asylum, 1845-1914, Medical History 42 (1998) 1-31.
56. The concept of total distance is discussed below. The dates are the census points for population purposes but admissions to the asylum are grouped around these dates-1845-54, 1855-64 and so on-for each census point. The effective patient intake period is 1845 to 1904.
57. This z-score technique compared favourably, in a test for robustness, with a simple correlation between numbers admitted and distance, which has remained the basis of many calculations of distance-decay since Jarvis.
58. The z-score formula was as follows: if the total population is TV and the total number of lunatics is M, then p = M/N; q = (1-). If the population of any particular subgroup is n, and the number of lunatics admitted from there is m then for any particular subgroup, the expected number of lunatics is np and the z-score is given by m -np/the square root of npq.
59. In principle, it might be expected that there would be a very even distribution of pauper lunatics across Devon given the assumptions of the z-score model. Our results indicate a significant range of experience between the different parishes. The distribution that the data are expected to obey by the laws of chance alone is the binomial; in this case a distribution, with mean np and variance npq. This is then approximated to the normal (0,1) distribution, an approximation which is generally deemed to be valid as long as np 5. To our knowledge, this is the first attempt at such an exercise using this kind of parish data.
60. The system used was that of ESRI Arcinfo. We acknowledge the assistance of Yu Zhang and Rob Ashmore for their help in digitising parish boundaries and advice on data processing. Distance is measured as the crow flies rather than by proximity to roads, railways and canals, as Jarvis attempted to do. It is difficult to engage with Jarvis's arguments on the impact of road, canal and rail networks on the distance effect given the greater density of road networks in English counties as compared to US states.
61. There are some technical objections in using z-scores of variance for regression exercises, though we remain confident that the results are to be preferred to a simple calculation of patient proportions within the respective parish populations.
62. It is tempting to speculate on the presence of genetic and family-specific characteristics in the absence of evidence though we have preferred to explore the influence of social and commercial structures on many of these high-scorers, as they included substantial commercial, market and trading centres such as the fishing port of Brixham, the ancient town of Ottery St Mary, and the booming area of Newton Abbot. Such prominent sending parishes were frequently within reach of the emerging tourist resorts along the Devon coastline in these decades.
63. Since it was extremely difficult to compensate for 'policy effects' as the borough authorities changed their policies and sent varying numbers to Exminster, we excluded both cities from most exercises.
64. This assumption is open to question given the location of numerous county asylums within a short distance of the main county town and other population centres where they would logically expect to draw most of their patients. The Devon Asylum was built within five miles of Exeter and adjacent to the main road and rail links to Plymouth. Including Plymouth and Exeter in the regression exercise changes the 'sign' for the physical area equation to positive and the results then indicate that large parishes were associated with admissions, though Plymouth and Exeter are clearly identified as 'outliers' and have markedly opposing signs in 1851 and Exeter as an outlier in every period thereafter. Such results suggest that the two cities have a distorting effect on the results and it is preferable to omit them. The major boroughs of Barnstaple and Tiverton appear to have displayed a low propensity to send lunatics in the later part of the period (see Figures 1 to 4 above), and Barnstaple (Tiverton much less so) appears prominently as one of the 'outliers' which was negatively associated with low admissions in our z-score calculations. Alternatively, if we make the (defensible) assumption that the majority of the workhouse residents were likely to be residents of the town in which the workhouse was located then the picture changes considerably.
65. The strong positive association between the 404 nearby parishes and the Asylum which emerged in 1851-71 continued amongst 583 nearby parishes sending in 1881-1901, but the negative distance effect weakened amongst those at a moderate distance from Exminster whilst the negative effect actually sharpened for the further parishes, and remained similar for the furthest 400-550 parishes. Once again there was no simple linear relationship between space and admission. The parish numbers are the cumulative totals for the three census periods surveyed.
66. The parishes with the smallest populations (less than 500) accentuated their negative sign in the period 1881-1901 as compared with 1851-71 from -4-5 to -5-7, whilst the slightly larger parishes (500-999 inhabitants) weakened from -5-1 to -3-6. The strongest positive association between middling-sized parishes continues but the effect 'filters down' over the two periods from those with 4000-6000 inhabitants to those with 2000-4000 residents. As noted above, including Plymouth and Exeter results in a strongly negative correlation between size of population and propensity to send, more particularly after '1881' (or after the period 1875-84). It seems likely that the small numbers from these major boroughs significantly affected the overall results when they were included. Excluding the smaller boroughs of Barnstaple and Tiverton from the regression yielded no significant change in the results reached for those unions. This suggests that the impact of alternative amenities by Exeter and Plymouth at Digby's Field in 1886 and Moorhaven in 1892, was greatest on those cities rather than on other Devon unions.
67. The published figures for all Devon unions indicate that neither Plymouth nor Exeter possessed a significantly larger number of recorded lunatics than the average. Both recorded substantial increases in 1881, though neither was as great as Honiton. The dramatic z-scores recorded in 1901 in Exeter was not matched by Plymouth, again suggesting data problems and/or local circumstances.
68. Very few towns possessed such populations, though they were growing steadily throughout the period probably at the expense of smaller settlements. Substantial towns such as Barnstaple were subject to a multiplicity of pressures influencing the disposal of pauper lunatics. Large numbers of known lunatics in Plymouth and Devon during the later nineteenth century were housed in workhouses and private establishments prior to the building of their own borough institutions, though they may not have been as significant a proportion of the urban population as in other Devon towns.
69. To undertake this exercise we divided the population of each parish by its physical area and multiplied the resulting figure by 20 million to make the numbers more manageable. Parishes were coded according to the size of their population.
70. Generally the least dense parishes remained under-represented in admissions throughout the period, though marginally less so in the later decades. The most dense parishes were more strongly associated with admissions in the earlier decades. The intermediate parishes were a little more associated with admissions in the later period as the presence of the mostdense settlements weakened.
71. D. Wright, Getting out of the asylum, Social History of Medicine 10 (1997) 136-55, offers a useful survey.
72. Philo, Journey, 149.
73. The scale ranged 0-8 in each direction with ranges from 0-0001-5-0 percent to more than 35 per cent plus or negative. Since population data was not available for 1921 we have not included 1911, though we did extract data for 1841 so the 1851 figure is provided.
74. The period was 1875-84 matched against the trend in population 1871-91. For every other year the test of significance shows very little reliance can be placed on the result. The regression result again provides only a rough indication of the significance of one variable amongst many. Our results indicate that this variable provides only a limited part of an effective explanation of admissions.
75. Towards the end of the period there is slight evidence that the most rapidly-growing parishes of all were associated with higher asylum admissions whilst those losing people the most quickly were rather poorly represented in admissions, though even this snapshot view is an impressionistic one.
76. R. Adair, J. Melling and B. Forsythe, Migration, family structure and pauper lunacy in Victorian England: admissions to the Devon County Pauper Lunatic Asylum, 1845-1900, Continuity and Change 12 (1997) 373-401.
77. A number of parishes within some Poor Law unions saw high rates of population growth and a strong rate of admission to the asylum at different periods in the later nineteenth century, though the Union share of Exminster admissions was often less marked. Newton Abbot Union enclosed the rapidly-growing areas of Torbay but in only one period (1885-1894) was there a strong association between residence in this union and asylum intake. Table 6 indicates the extent to which Newton Abbot relied on a range of institutional and non-institutional means to house its lunatics in 1871-1901 and its low z-scores for union lunatics in these years of rapid population growth.
78. A contrast with Newton Abbot and south Devon can be found in Barnstaple Union on the north Devon coast where Barnstaple town itself made only limited use of the County Asylum but the regression results suggest that membership of this Poor Law Union had either no significant, or a mildly positive, influence on admissions to the asylum. Such returns again indicate the complexities of demographic and institutional effects in play.
79. Adair, Forsythe and Melling, Migration, family structure and pauper lunacy.
80. The authors are presently engaged in a detailed analysis of the machine-readable 1881 census enumerators' data provided by the Essex Data Archive.
81. The results from the multiple regression for 1851, for example, suggested an R Square of 0-13142 and an adjusted R Square score of 0-12307.
82. As noted distance was measured between the different Devon parishes and the centre of the parish of Exminster (the approximate location of the Devon County Asylum), rather than as the traveling distance by road, rail or sea voyage. This estimate was combined with a measurement of the distance between the particular parish and the centre of the Poor Law union which was taken as the location of the workhouse. In this sense we tried to capture some aspects of the 'administrative distance' we discuss below. We have not found clear evidence of any 'zone of indifference' within Devon.
83. J. Melling, B. Forsythe and R. Adair, Families, communities and the legal regulation of lunacy in Victorian England. Assessments of crime, violence and welfare in admissions to the Devon Asylum 1845-1914, in P. Bartlett and D. Wright (Eds), Outside the Walls of the Asylum: The History of Care in the Community 1750-2000 (Athlone 1999) 153-80, 304-11.
84. Some known lunatics were boarded in houses maintained by carers who were neither relatives nor friends but who acted on monetary incentives offered by the Poor Law authorities, families and others who were keen to retain the individual in 'the community' rather than a specialist or Poor Law institution.
85. This means that a direct comparison with our data on the admissions to the Devon Asylum, which are constructed from the ten years around the census year, is not possible.
86. The data for census years prior to 1871 was too fragmentary and inconsistent to utilize for our analysis of institutional policies of the Devon unions.
87. Independent boroughs did not contribute to the construction costs of the County Asylum and possesed no right of access, being charged a higher rate for the entry of their pauper lunatics.
88. There were a total of 120 known pauper lunatics in such houses in Devon itself in 1876, rising to 281 in 1886. The only noticeable presence was that of Newton Abbot, which used both private mad houses and the workhouse to a significant degree in the period 1876-1906, probably in response to overcrowding at the County Asylum. All data was derived from the Annual Reports of the Lunacy Commissioners.
89. A total of approximately 300 lunatics were recorded in Devon workhouses until the end of the century, with numbers falling below 200 in 1901. Most Boards of Guardians responded to pressures from the Lunacy Commission by making less use of workhouses from the 1880s, though the flight from workhouse accommodation was a slow retreat rather than a sudden rush. These numbers include Exeter and Plymouth.
90. There appear to have been about 300 Devon lunatics cared for in 'the community' in 1871 and a little over 200 in 1901. This is equivalent to almost half the number held at Exminster in 1871 but is the equivalent of only one fifth by 1901. As noted above, it is likely that a proportion of these individuals were boarded out to non-relatives either on a direct commercial basis or by arrangement with the Poor Law Guardians, underlining the need for a careful assessment of arguments that Asylum places were created in response to 'community reaction'.
91. The Lunacy Commission data included East Stonehouse and Stoke Damerel (two of the three 'towns' of Plymouth which each comprised a distinct union for Poor Law purposes), as well as Exeter and Plymouth. The returns given in these Reports frequently vary in some respects from our results based on the analysis of Exminster admission registers, probably due to the exclusion of 'workhouse' inmates from our data where the parish of origin is unclear and also from the inclusion in Commission Reports of all asylum inmates within one particular union, rather than those in the Devon County Asylum. The data from the Lunacy Commission Reports is readily useable only from the 1860s and there remain technical problems in using the figures for comparisons. The figures we derived from the Reports on Devon covered only the census years, whereas our Exminster data covers the decade 'around' the census year (e.g. 1845-54 admissions and census population for 1851). To check the results we also calculated data for the mid-decennial point of 1866, 1876, etc. which yielded similar results to those derived from decennial years.
92. To remove the risk of obvious bias in that the Poor Law physicians and workhouse officers administered the certification process and arrivals at Exminster were frequently identified as coming from the workhouse rather than a parish, we omitted the obvious workhouse residents and those whose parish of origin was unclear. A intra-union distance effect remained visible.
93. Forsythe, Melling and Adair, Politics of lunacy.
94. As noted above the modern city of Plymouth is composed of three ancient 'towns', namely Plymouth, Stoke Damerel (later Devonport), and East Stonehouse (later Stonehouse), which are not included in Table 6. Broadly, we find that East Stonehouse recorded markedly low numbers of recorded lunatics in Devon whilst Plymouth itself possessed large numbers of lunatics in 1881 and again in 1901, with many houses in private asylums and the workhouse, though a dramatic drop in its constituents in private mad-houses in 1891 (probably Plympton House) was responsible for the fall to insignificance at that point, prior to the opening of the Moorhaven Asylum for the Borough in 1892. As with other boroughs, it is likely that significant numbers were sent outside the county when county facilities were unavailable. Exeter saw a dramatic increase in the total number of pauper lunatics in 1881-1901, mostly housed in its borough asylum after 1886, though with significant numbers housed with friends and relatives and in its workhouse even in the 1890s. The evidence points to the need for more detailed research on urban lunacy provisions in this period.
95. Adair, Forsythe and Melling, A danger to the public?
96. B. Forsythe, J. Melling and R. Adair, The New Poor Law and the county pauper lunatic asylum: the Devon experience, 1834-1884, Social History of Medicine 9 (1996) 335-56, provides a preliminary assessment.
97. We further tested the significance of the union effect by replacing the results of the St Thomas and Kingsbridge unions with those for Newton Abbot, a union with a steadily rising population, and Barnstaple (which included a borough which used the asylum only lightly at different periods). The regression results gave a slightly improved 'fit' as indicated by the R-squared return at some points but there was little variation from the results using the first two unions. This would imply that little different arises from using different union data and that union-wide effects have a limited impact on the results.
98. Philo, Journey to the asylum, 159-60, where various private facilities are discussed. For reasons of space we have not discussed the existence of private mad-houses in Devon. We have noted that Plymouth used the notorious establishment at Plympton for many years and the Exeter Guardians sent their pauper lunatics to Somerset and Dorset institutions. Philo suggests is that the existence of such profit-making establishments helps to explain the limited distance-decay effect in English counties, as the more affluent families sent their relatives to these places rather than the pauper asylums. There appears little evidence in Devon at least that private mad-houses significantly affected the distance-effect pattern we have outlined.
99. For these institutions see W. L. Parry Jones, The Trade in Lunacy. A Study of Private Madhouses in England in the Eighteenth and Nineteenth Centuries (London 1972)
100. and C. MacKenzie, Psychiatry for the Rich: A History of Ticehurst Private Asylum, 1792-1917 (London 1992). The whole question of social position, resources and the use of pauper institutions justifies detailed reappraisal in this respect.