1 Gloucestershire Record Office (hereafter GRO), Gloucestershire Wills, 1551/16.

2 Lambeth, Register, Cranmer, fo. 105, fos. 332–3.Google Scholar

3 ‘Bishop Hooper's Visitation Booke’ in Dr Williams's Library, Morice MS 31.L, Item 3, 16. This was the last will in Edward's reign to contain the standard traditional preamble, based on a reading of over 3,000 Gloucestershire wills for the period from 1541 to 1580.

4 GRO, Gloucestershire Wills, 1551/62.

5 The Testament of Master Wylliam Tracie esquier / expounded both by William Tyndale and Jhon Frith (Antwerp, 1535), STC 24167; ‘The Testament of W. Tracie expounded by W. Tindall’, in Wyclyffes Wycket: whyche he made in kyng Rycards days the second in theyere ofourlorde God M.CCC.XLV (1546), STC 25590; ‘The Testament of W. Tracie expounded by W. Tindall and J. Frythe’, in Uvicklieffes Wicket. Faythfully ouerseene and corrected (1548), STC 25591; Concilia Magnae Britanniae et Hiberniae ab anno MDXLVI ad annum MDCCXVII, 4 vols., ed. Wilkins, A. (1737), vol. 3, 746–7Google Scholar. See also John, Craig and Caroline, Litzenberger, ‘Wills as religious propaganda: the testament of William Tracy’, Journal of Ecclesiastical History (hereafter JEH) 44 (1993), 415–31.Google Scholar

6 In this article the terms, ‘the old religion’ and, from Mary's reign on, ‘Catholicism’, will be used synonymously with ‘traditional religion’ to describe the standard pre-Reformation religion itself, and its subsequent forms. The terms, ‘protestantism’ and ‘the new religion’, will be used interchangeably to describe the forms of religion introduced by the Reformation.

7 Haigh, C., ‘The recent historiography of the English Reformation’, Historical Journal 25 (1982), 995–1007CrossRefGoogle Scholar; Dickens, A. G., ‘The early expansion of protestantism in England, 1520–1558’, Archiv fur Reformationsgeschichte 78 (1987), 187–222.Google Scholar

8 Elton, G. R., Reform and Reformation: England, 1509–1558 (Cambridge, MA, 1977), 371Google Scholar. See also Dickens, A. G., The English Reformation (2nd edn, London, 1989).Google Scholar

9 Haigh, C., Reformation and resistance in Tudor Lancashire (Cambridge, 1975)Google Scholar; Scarisbrick, J. J., The Reformation and the English People (Oxford, 1984).Google Scholar

10 While the terms ‘will’ and ‘testament’ legally refer to two different instruments, they are used synonymously in this article. For a discussion of the distinctions between the terms see Pollock, F. and Maitland, F. W., The history of English law before the time of Edward I (2nd edn, 1898; rep. Cambridge, 1968), vol. 2, 314–56Google Scholar; Coppel, S., ‘Wills and the community of Tudor Grantham’, in Probate records and the local community, ed. Riden, P. (Gloucester, 1985), 72–4Google Scholar. For further discussion of wills, probate and testamentary cases in the consistory court, see Houlbrooke, R. A., Church courts and the people during the English Reformation, 1520–1570 (Oxford, 1979), 89–116, passim.Google Scholar

11 Approximately 8,000 Gloucestershire wills proved in the diocesan consistory court survive in the archives of the Gloucestershire Record Office for the period of this study. In addition to all extant wills of female testators, 20 per cent of the wills of lay men were included. Since the wills were filed alphabetically by first name within probate year, it was possible to select the sample of male wills by identifying the first man's will in each year based on a random number (modular 5), and then reading every fifth man's will through the rest of that year. The resulting sample includes approximately equal numbers of women's and men's wills. In this case, assuming the highest possible variability of the sample from the population, the results of the analysis of the sample will be within 0.97 per cent of that for the entire population at a confidence level of 95 per cent. Further, chi-square tests, which measure the strength of relationships and the importance of differences, have been used to identify important patterns in groups of wills, at a significance level of 0.001. See Schofield, R. S., ‘Sampling in historical research’, in Nineteenth-century society: essays in the use of quantitative methods for the study of social data, ed. Wrigley, E. A. (Cambridge, 1972), 146–84, passimCrossRefGoogle Scholar; Blalock, H. M., Social statistics (rev. edn, London, Tokyo, Paris and Singapore, 1981), 183–6Google Scholar; and Floud, R., An introduction to quantitative methods for historians (2nd edn, 1973; rep., London and New York, 1986), 136.Google Scholar

12 Spufford, M., ‘The scribes of villagers' wills in the sixteenth and seventeenth centuries and their influence’, Local Population Studies (hereafter LPS) 7 (1972), 28–43.Google Scholar

13 Dickens, A. G., Lollards and Protestants in the diocese of York (1959; repr., London, 1982), 221.Google Scholar

14 Matlock Population Studies Group, ‘Wills and their scribes’, LPS 8 (1972), 55–7Google Scholar; Richardson, R. C., ‘Wills and will-makers in the sixteenth and seventeenth centuries: some Lancashire evidence’, LPS 9 (1972), 33–42Google Scholar; Capp, B., ‘Will formularies’, LPS 14 (1975), 49Google Scholar; Zell, M. L., ‘The use of religious preambles as a measure of religious belief in the sixteenth century’, The Bulletin of the Institute of Historical Research 50 (1977), 246–9CrossRefGoogle Scholar; Vann, R. T., ‘Wills and the family in an English town: Banbury, 1550–1800’, Journal of Family History 4 (1979), 346–67CrossRefGoogle Scholar; Attreed, L. C., ‘Preparation for death in sixteenth-century Northern England’, Sixteenth Century Journal 13 (1982), 37–66CrossRefGoogle ScholarPubMed; Mayhew, G. J., ‘The progress of the Reformation in East Sussex 1530–1559: the evidence from wills’, Southern History 5 (1983), 38–67Google Scholar; Whiting, R., ‘“For the health of my soul”: prayers for the dead in the Tudor South-west’, Southern History 5 (1983), 68–94Google Scholar; Cross, C., ‘Wills as evidence of popular piety in the Reformation period: Leeds and Hull’, in The End of Strife, ed. Loades, D. (Edinburgh, 1984), 44–51Google Scholar; Evans, N., ‘Inheritance, women, religion and education in early modern society as revealed by wills’, in Probate records and the local community, ed. Riden, P. (Gloucester, 1985), 53–70Google Scholar; Coppel, , ‘Wills and the community’ 71–90Google Scholar; Cressy, D., ‘Kinship and kin interaction in early modern England’, Past and Present 113 (1986), 38–69CrossRefGoogle Scholar; Cross, C., ‘Northern women in the early modern period: the female testators of Hull and Leeds 1520–1650’, Yorkshire Archaeology Journal 59 (1987), 83–94Google Scholar; Alsop, J. D., ‘Religious preambles in early modern English wills as formulae’, JEH 40 (1989), 19–27Google Scholar; Burgess, C., ‘Late medieval wills and pious convention: testamentary evidence reconsidered’, in Profit, piety and the professions in later medieval England, ed. Hicks, M. A. (Gloucester, 1990), 14–33Google Scholar; Marsh, C., ‘In the name of God? Will-making and faith in early modern England’, in The records of the nation, ed. Martin, G. H. and Spufford, P. (Woodbridge, 1990), 215–49Google Scholar; Prior, M., ‘Wives and wills, 1558–1700’, in English Rural Society, 1500–1800, eds. Chartres, J. and Hay, D. (Cambridge, 1990), 201–25Google Scholar; Takahashi, M., ‘The number of wills proved in the sixteenth and seventeenth centuries. Graphs, with tables and commentary’, in Martin, and Spufford, eds., The records of the nation, 187–213Google Scholar; McIntosh, M. K., A community transformed: the manor of Havering, 1500–1620 (Cambridge, 1992), 85–91, 188–94, passimGoogle Scholar; Duffy, E., The stripping of the altars: traditional religion in England, c. 1400–c. 1580 (New Haven, CT, and London, 1992), 504–23, passimGoogle Scholar; and Craig, and Litzenberger, , ‘Wills as religious propaganda’.Google Scholar

15 Marsh, , ‘In the name of God?’, 219, 238, 243.Google Scholar

16 The wills data were collected in as much detail as possible, including verbatim transcriptions of the preambles. Subsequently, a set of codes was developed based on the textual content analysis to describe every distinctive soul bequest found in the wills that had been read, and the coded preambles were categorized. Then, statistical analysis was performed by examining the incidence of the various preamble categories as percentages of the whole.

17 This is similar to the methodology employed by Marjorie McIntosh in the identification of will scribes, although her approach differs somewhat due to differences in the information available for analysis. See McIntosh, , A community transformed, 88.Google Scholar

18 GRO, Gloucestershire Wills, 1559/216.

19 GRO, Gloucestershire Wills, 1555/29.

20 GRO, Gloucestershire Wills, 1558/223.

21 GRO, Gloucestershire Wills, 1571/18.

22 GRO, Gloucestershire Wills, 1573/93.

23 GRO, Gloucestershire Wills, 1575/124, 1577/80.

24 GRO, Gloucestershire Wills, 1554/78.

25 GRO, Gloucestershire Wills, 1560/103.

26 In Figure 1 and Table 3, with a chi-square value of 525.95 with 8 degrees of freedom, a very significant value. This value was determined by analyzing preamble counts by reign or portion of reign within the general categories, traditional, neutral and protestant, and comparing the differences in the preference for each over time with that which would have been expected based on the overall percentage of wills with preambles in each general category. The reigns or portions of reigns include the last years of the reign of Henry VIII, the complete reigns of Edward VI and Mary I, and the first half of the reign of Elizabeth I, with the latter divided into two separate portions for analysis purposes: Henry VIII (1541–1547), Edward VI (1547–1553), Mary I (1553–1558), Elizabeth I (1559–1569, and 1570–1580). These periods are used throughout the analysis described in this article.

27 The parish provisions in Gloucestershire for liturgical conformity to Marian Catholicism will be discussed in my Cambridge University Ph.D. thesis, ‘Responses of the laity to changes in official religious policy in Gloucestershire (1541–1580)’, chapter 5.

28 GRO, Gloucestershire Wills, 1556/64.

29 GRO, Gloucestershire Wills, 1557/224.

30 In comparing the use of neuiral preambles by men and women there is a chi-square value of 26.70 with 4 degrees of freedom. The chi-square value for neutral preambles was determined by tabulating those particular preambles by gender and by reign or portion of reign. Then the actual preferences for neutral statements in women's and men's wills over time were compared to determine if there was a significant difference by gender. The test for protestant wills is probably not statistically significant due to the small numbers of wills involved.

31 Of the 647 Elizabethan men who left wills, 275 (42.5%) included either their status or occupation. It is not possible to analyze Elizabethan women's wills in this way, because most merely gave their status as ‘widow’ without further elaboration, and their wills seldom included material evidence of their social position or wealth.

32 Fewer of the aldermen and gentry chose neutral statements, but with only nine wills in that segment of the sample, their preferences can not be considered representative.

33 Chi-square value = 46.05 with 16 degrees of freedom. This value was determined by comparing the preferences for neutral wills over time in each region, having accumulated the will-counts by reign or portion of reign.

34 The funds bequeathed by Margaret Hyckes were to be administered by the parish churchwardens; GRO, Gloucestershire Wills, 1562/71.

35 GRO, Gloucestershire Wills, 1569/156, 1574/143, 1579/107.

36 Collinson, P., The birthpangs of protestant England: religious and cultural change in the sixteenth and seventeenth centuries (Basingstoke, 1988), p. ix.Google Scholar