¹ Stenton, F. M., The First Century of English Feudalism, 1066–1166, (2nd ed.; Oxford, 1961), p. 142, n. 4.Google Scholar

² See, for example, Hilton's, R. H. similarly informal, and similarly correct remark on the subject in A Medieval Society: the West Midlands at the End of the Thirteenth Century (New York, 1966), p. 53.Google Scholar

³ For instance, Bumke, Joachim on the term Ritter in The Concept of Knighthood in the Middle Ages, trans. Jackson, W. T. H. and Jackson, Erika (New York, 1982)Google Scholar; Flori, Jean on chevalier in “La notion de Chevalerie dans les Chansons de Geste du XII^e siècle: Etude historique de vocabulaire,” Moyen Age 81 (1975): 211–44, 407–45Google Scholar; Georges Duby on miles in (among many others) “The Origins of Knighthood,” in The Chivalrous Society, trans. Postan, Cynthia (Berkeley, 1977), pp. 158–70Google Scholar. The only full-scale study along these lines for England is Delehanty's, William M. doctoral dissertation, “Milites in the Narrative Sources of England, 1135–54” (University of Minnesota, 1975)Google Scholar, though Richard Abels provides a useful survey of Anglo-Saxon and early Anglo-Norman milites in chapter seven of Lordship and Military Obligation in Anglo-Saxon England (Berkeley, 1988)Google Scholar. For a useful review of the literature that contrasts work on English and French knights, see Hunt, Tony, “The Emergence of the Knight in France and England, 1000–1200,” Forum for Modern Language Studies 17 (1981): 93–114.CrossRefGoogle Scholar

⁴ See, for example, Evergates, Theodore, Feudal Society in the Baillage of Troyes under the Counts of Champagne, 1152–1284 (Baltimore, 1975)Google Scholar, and Duby, Georges, “Lineage, Nobility and Knighthood: the Mâconnais in the Twelfth Century – a Revision,” in The Chivalrous Society, pp. 59–80Google Scholar. For a discussion of much recent literature on the topic, see Flori, Jean, L'essor de la chevalerie, XI^e-XII^e siècles (Geneva, 1986), pp. 9–38Google Scholar. The word class appears in this study only with its basic meaning of set or group; no implication that knights did (or did not) constitute a social or economic class (however defined) is intended.

⁵ Reginald Treharne gave the most ebullient statement of this common thesis; “The Knights in the Period of Reform and Rebellion, 1258–67,” in Simon de Montfort and Baronial Reform: Thirteenth Century Essays, ed. Fryde, E. B. (London, 1986), pp. 268–71.Google Scholar

⁶ All of the documents from three sources were inspected for possible inclusion in the database: Round, John H., ed. Ancient Charters, Royal and Private, Prior to A.D. 1200, Pipe Roll Society, o.s., vol. 10 (London, 1888)Google Scholar; Loyd, Lewis C. and Stenton, Doris M., eds., Sir Christopher Hatton's Book of Seals (Oxford, 1950)Google Scholar; Patterson, Robert B., ed., Earldom of Gloucester Charters (Oxford, 1973)Google Scholar. Only a limited range of the acta from the four other sources were examined: Greenway, Diana E., Charters of the Honour of Mowbray 1107–91, British Academy Records of Social and Economic History, n.s., vol. 1 (London, 1972)Google Scholar [through no. 84]; Mason, Emma, ed., The Beauchamp Cartulary Charters, 1100–1268, Pipe Roll Society, n.s., vol. 43 (London, 1980) [nos. 1–78, 252–69]Google Scholar; Dodwell, Barbara, ed., The Charters of Norwich Cathedral Priory, Part Two, Pipe Roll Society, n.s., vol. 46 (London, 1985) [nos. 7–201]Google Scholar; Kemp, B. F., ed., Reading Abbey Cartularies, Camden Society, 4th ser., vols. 31, 33 (London, 1986–1987), [nos. 236–818].Google Scholar

⁷ This restriction did not apply to Hatton 's Book of Seals, as it is itself a compendium from a number of sources.

⁸ Royal writs tend to have one one or two witnesses; Van Caenegem, R. C., Royal Writs from the Conquest to Glanvill, Seldon Society, no. 77 (London, 1959), pp. 146–58.Google Scholar

⁹ For purposes of tabulation, every act has been assigned to one of ten twenty-year blocks covering the years 1101–1300 (i.e., 1101–20, 1121–40, etc.). For acta printed with a numerical date range (i.e., year X to year Y), the mean of the range determines in which block the charter falls. For acts given qualitative dates the following conventions apply: early in a century = the first block (01–20), late = the last (81–100); mid-century = the block 41–60; early Henry II = 1141–60; Henry II = 1161–80; late Henry II = 1181–1200; John = 1201–20; late John/early Henry III = 1201–20; early Henry III = 1221–40; Henry III = 1241–60. These conventions may of course distort the distribution of acts over time; this problem becomes less acute the farther one advances into the thirteenth century, as more and more charters bear a precise written date.

¹⁰ For statistical analysis, descriptors do not include toponyms, patronyms, and other marks of kinship (including junior). More difficult are those witnesses who are apparently given both first and last name in the vernacular (e.g., Ralph Basset). These second names have not been treated as descriptors unless they indicate an occupation or status. In general, the terms used to describe men in these lists are remarkably monotonous. Leaving aside dominus and miles, the great majority of words appear to fall into one of the following categories: titles of nobility (e.g. comes), household or royal offices (dispensator, vicecomes), terms denoting rank within the church or membership in the clerical order (episcopus, clericus, magister), and names of trades or occupations (pistor, cocus).

¹¹ Duby found that early acts of the Mâconnais used the term nobilis (and equivalents like clarissimus) to describe beneficiaries, but not grantors, of leases—apparently a convention of diplomatic; “Lineage, Nobility and Knighthood,” p. 75.

¹² Of the 839 acta in the database, 709 (85%) apply descriptors (as defined above) to one or more witnesses. The highest proportion of charters using descriptors in any twenty-year block is 96% (100 of 104 charters) for 1261–80; the lowest is 76% (94 of 123) for 1181–1200.

¹³ Clanchy, Michael T., From Memory to Written Record, England 1066–1307 (London, 1979), pp. 34–35.Google Scholar

¹⁴ Restricting attention to acts employing descriptors, in the years 1101–1220, 3 of the 142 comital acta and 5 of the 267 non-comital ones employ miles to describe witness. For the period 1221–1300, the corresponding figures are 9 of 24 comital documents, and 106 of 276 non-comital.

¹⁵ Hatton's Book of Seals, no. 411.

¹⁶ For an example of witnesses being left out to make the list shorter, see Gloucester Charters, no. 84. Cartularies often omit the witness list altogether, but such copies have naturally not figured in this study. Differences in the spelling of names between copy and original, or among various copies, are of course common; ibid., nos. 31, 35, 95. See also Stringer, Keith, “The Charters of David, Earl of Huntingdon and Lord of Garioch: a Study in Anglo-Scottish Diplomatic,” in Essays on the Nobility of Medieval Scotland (Edinburgh, 1985), p. 92.Google Scholar

¹⁷ Three-quarters (36 of 48) of the acts using descriptors from 1281–1300 are originals. For 1221–1300 as a whole, 172 of the 300 acts with descriptors are originals (57%). The facsimiles and transcripts which make up Hatton's Book of Seals have been classed among the originals, since they were made for antiquarian purposes, and where they can be checked against surviving originals are very accurate; see the introduction, p. xv.

¹⁸ That is, 69 of the 172 original acts employing descriptors, versus 46 of the 128 copies.

¹⁹ Of charters with descriptors from the years 1281–1300, 10 out of 12 copies (83%) employ miles, while only 22 of 36 originals (61%) do so.

²⁰ Russell, Josiah C., “Attestation of Charters in the Reign of John,” in Twelfth Century Studies (New York, 1978), pp. 219–20Google Scholar. See also Greenway, Diana, “Ecclesiastical Chronology: Fasti 1066–1300,” in Baker, Derek, ed., The Materials, Sources and Methods of Ecclesiastical History (Oxford, 1975), pp. 57–60.Google Scholar

²¹ Of the 84 charters from the years 1201–1300 that mention more than one miles as witness, only two repeat the word rather than use the convention described above (Beauchamp Cartulary, no. 74, Hatton's Book of Seals, no. 329).

²² Since rank in witness lists is counted from the head of the list downward, the closer the depth is to 0, the closer the final miles is to the beginning of the list.

²³ Of the 116 thirteenth-century charters using miles to describe witnesses, 66 (57%) have a depth of 30% or less and 89 (77%) have a depth of 40% or less.

²⁴ Halton's Book of Seals, nos. 48, 138; Earldom of Gloucester Charters, no. 5, Reading Abbey Cartularies, no. 304.

²⁵ For vicecomes, Beauchamp Cartulary, no. 58; for magister and persona, see Hatton 's Book of Seals, nos. 15, 411, 479; for rector (ibid., no. 230); for justiciar regis (ibid., no. 56); for senescallus (ibid., nos. 19, 97, 336, 479). At least some of the witnesses entitled milites appear below men described as vicecomes, magister or persona in other charters (ibid., nos. 11, 189. 234).

²⁶ For example, lords of manors, high church officials, husbands, even the king; see Latham, R. E., Revised Medieval Latin Word List (London, 1965)Google Scholar, s.v. dominus. As Leopold Genicot notes, the power to command was a fundamental characteristic of the medieval aristocracy; “Recent Research on the Medieval Nobility,” in Reuter, Timothy, ed., The Medieval Nobility (Amsterdam, 1979), pp. 23–24.Google Scholar

²⁷ Since Table C includes only charters bearing descriptors other than miles, the increasing usage of dominus is not merely a function of the rise in popularity of the other word.

²⁸ In positioning the word dominus before the name, scribes imitated the vernacular, in which sieur preceded the name of persons so described. When placed between a name and a toponym, dominus apparently denoted lordship over the place (Evergates, , Feudal Society, pp. 100–101Google Scholar). For an instance of the latter practice, see Hatton's Book of Seals, no. 507.

²⁹ Thirty-seven of the charters do not employ the plural to designate domini; 32 of these clearly accord one or more men both titles. Of the 54 documents using the plural, 32 separate the word dominis from militibus by three names or less, implying a degree of overlap.

³⁰ Beauchamp Cartulary, no. 74.

³¹ Treharne, , “The Knights in the Period of Reform,” pp. 269–74Google Scholar; Hilton, , A Medieval Society, pp. 52–53Google Scholar. This theme of the growing restriction of the chivalric class is also the thesis of Harvey's, Sally “The Knight and Knight's Fee in England,” though she puts the decisive shift from professional to gentry knights in the first half of the twelfth century (Past and Present 49 [1970]: 40–43).CrossRefGoogle Scholar

³² Denholm-Young, Noel, “Feudal Society in the Thirteenth Century: the Knights,” in Collected Papers (Cardiff, 1969), pp. 83–87.Google Scholar

³³ Hilton's research in the West Midlands led him to doubt the estimate, for he found approximately 200 families in three counties which produced knights in the second half of the thirteenth century—considerably more than Denholm-Young's figure would allow (A Medieval Society, pp. 54–55). Saul, Nigel attacked Hilton's, calculations in Knights and Esquires: The Gloucestershire Gentry in the Fourteenth Century, (Oxford, 1981), p. 35Google Scholar. In general, scholars have seemed unwilling to question Denholm-Young's estimate: although J. Quick found a total of 1,539 knights by examining thirteenth-century grand assize jurors for 27 of the 39 counties of England, he concluded that “the order of magnitude of knightly numbers in the thirteenth century envisaged by Denholm-Young seems, therefore, to be roughly correct” (“The Number and Distribution of Knights in Thirteenth Century England: the Evidence of the Grand Assize Lists,” in Coss, P. R. and Lloyd, S. D., eds., Thirteenth-Century England (Woodbridge, Suffolk, 1986), p. 119Google Scholar).

³⁴ Beeler, John, Warfare in England, 1066–1189 (Ithaca, 1966), pp. 265–66.Google Scholar

³⁵ Hilton, , A Medieval Society, pp. 49–55Google Scholar; Postan, Michael M., The Medieval Economy and Society (Harmondsworth, Middlesex, 1975), pp. 178–84Google Scholar; Coss, P. R., “Sir Geoffrey de Langley and the Crisis of the Knightly Class in Thirteenth-Century England,” Past and Present 68 (1975): 3–37.CrossRefGoogle Scholar

³⁶ King, Edmund, “Large and Small Landholders in Thirteenth-Century England: the Case of Peterborough Abbey,” in Aston, T. H., ed., Landlords, Peasants and Politics in Medieval England (Cambridge, 1987) pp. 145, 163–64Google Scholar; Carpenter, D. A., “Was there a Crisis of the Knightly Class in the Thirteenth Century? The Oxfordshire Evidence,” English Historical Review 95 (1980): 724, 751.Google Scholar

³⁷ Raban, Sandra, “The Land Market and the Aristocracy in the Thirteenth Century,” in Greenway, Diana, Holdsworth, Christopher and Sayers, Jane, eds. Tradition and Change: Essays in Honour of Marjorie Chibnall (Cambridge, 1985), p. 261Google Scholar; Carpenter, , “Was there a Crisis of the Knightly Class?” p. 751.Google Scholar

³⁸ Ibid., pp. 721–52. Unfortunately, Carpenter's investigation covers the years 1220–1300; witnesslist data and the chronology of inflation imply that the narrowing of the chivalric class was largely accomplished by then. Carpenter's work does reflect some minor constriction of the chivalric class at the bottom. He divides the 57 families he traces into four categories, based on size of tenements. Only one or two of the seven families of his lowest category (those that held less than three hides of land) survived the century successfully.

³⁹ Palmer, Robert C., “The Economic and Cultural Impact of the Origins of Property, 1180–1220,” Law and History Review 3 (1985): 381–89.CrossRefGoogle Scholar

⁴⁰ For the second possibility, see Trehame, , “Knights in the Period of Reform,” p. 275.Google Scholar

⁴¹ It is not certain that the cost of a knight's equipment increased substantially between 1100 and 1220 (Brown, R. Allen, “The Status of the Norman Knight,” in Gillingham, John and Holt, J. C., eds., War and Government in the Middle Ages: Essays in Honour of J. O. Prestwich [Cambridge, 1984], p. 31Google Scholar).

⁴² Contamine, Philippe, War in the Middle Ages, trans. Jones, Michael (Oxford, 1984), p. 69.Google Scholar

⁴³ Powicke, Michael, Military Obligation in Medieval England: a Study in Liberty and Duty (Oxford, 1962), pp. 54–56, 85–89, 119–20.Google Scholar

⁴⁴ The theory that the Crown enforced arms and armor obligations more regularly against knights than others might account for the royal interest in forcing men to become knights (i.e., distraint of knighthood). On distraint, see Powicke, , “Distraint of Knighthood and Military Obligation under Henry III,” Speculum 25 (1950): 457–70CrossRefGoogle Scholar; Waugh, Scott L., “Reluctant Knights and Jurors: Respites, Exemptions and Public Obligations in the Reign of Henry III,” Speculum 58 (1983): 937–86.CrossRefGoogle Scholar

⁴⁵ Evergates, , Feudal Society, pp. 114–15, 145–50Google Scholar; Beech, George T., A Rural Society in Medieval France: The Gâtine of Poitou in the Eleventh and Twelfth Centuries (Baltimore, 1964), pp. 95, 101Google Scholar; Duby, Georges, “The Nobility in Medieval France,” in The Chivalrous Society, p. 108Google Scholar. Léopold Génicot would put the change some twenty years later for Namur; Les hommes — la noblesse, vol. 2 of L'économie rurale Namuroise au Bas Moyen Age (Louvain, 1960), pp. 122–23.Google Scholar

⁴⁶ Duby, , “The Transformation of the Aristocracy,” pp. 178–80Google Scholar. For a more cautious viewpoint on the assimilation of nobles and knights in France, which reviews much of the literature of the last thirty years, see Flori, Jean, L'essor de la chevalerie, pp. 21–36, 119–141.Google Scholar

⁴⁷ Milsom, S. F. C., The Legal Framework of English Feudalism (Cambridge, 1976), especially pp. 8–14, 25–42CrossRefGoogle Scholar; Palmer, Robert C., “The Feudal Framework of English Law,” Michigan Law Review 79 (1981): 1132–53CrossRefGoogle Scholar, and “The Origins of Property in England,” Law and History Review 3 (1985): 1–8, 18–24.Google Scholar

⁴⁸ For the decline of the old servitium, see Sanders, Ivor J., Feudal Military Service in England (London, 1956), pp. 59–90.Google Scholar

⁴⁹ Waugh, Scott L., “Tenure to Contract: Lordship and Clientage in Thirteenth–Century England,” English Historical Review 101 (1986): 811–39.CrossRefGoogle Scholar

⁵⁰ For an evaluation of the wedge the Common Law drove between a free tenant's relation with his lord and a villein's relation to the lord of the manor, see Searle, Eleanor, “Seigneurial Control of Women's Marriage: The Antecedents and Function of Merchet in England,” Past and Present 82 (1979): 8–18, 34–35.CrossRefGoogle Scholar

⁵¹ So David Crouch suggested in the Constance Egan lecture to the Heraldry Society, Society of Antiquaries, November 18, 1987.

⁵² Treharne, , “The Knights in the Period of Reform,” pp. 271–72Google Scholar. For a useful summary of the administrative work of knights, see Poole, Austin Lane, Obligations of Society in the XII and XIII Centuries (Oxford, 1946), pp. 53–56.Google Scholar

⁵³ I must thank Richard Abels for calling to my attention the possible link between the declining military role of the enfeoffed knight and the use of the miles in acta. For chivalric ideology, see Flori's Le essor de la chevalerie, especially chapters 12 and 14.