The_Composition_of_Famuli_Labour_on_Engl

The composition offamuli labour on English demesnes, c.1300*

by Jordan Claridge and John Langdon

Abstract

This article explores the nature of agricultural labour in England c.1300. Using a national sample of over 400 manorial accounts containing detailed data for over 4000 individuals, the piece looks closely at famuli labour, the nucleus of the workforce on seigneurial demesnes (the farms directly cultivated by manorial lords as opposed to the land of their tenants) at the beginning of the fourteenth century, a period considered to be the pinnacle of medieval population and intensive land exploitation. By examining the rates of remuneration as well as the availability of work for the range offamuli labourers, we argue that famuli labour was divided into a bipartite system of first- and second-tier workers where core, or first-tier (and mostly male), labourers such as ploughmen, carters, and shepherds were paid higher wages and presented with more opportunities to work as compared to a group of more subsidiary ‘second-tier’ labourers largely comprised of women, the young and the elderly.

This article is an exercise in examining the labour employed on medieval English demesnes

– the working farms of lords on manors as opposed to the lands of their tenants – in a more systematic, comprehensive, and innovative fashion than is available in the literature to date. We do this particularly to assess the numerical and other relationships between the routinely hired supervisory personnel, ploughmen, carters, shepherds, and so on – ‘first-tier’ labour, as we style it in this article and the more subsidiary or ‘second-tier’ labour that mostly existed to support and extend the effectiveness of the first-tier personnel. This latter, ancillary group encompassed much of the work of women, and certainly that of the young, the elderly, the poor, and perhaps even the disabled, in demesne workforces. A ‘snapshot’ of these various workers in demesne agriculture, referred to in totality as the famuli in the records, is provided in this study for around the beginning of the fourteenth century. This is the moment considered to be the peak of medieval population and intensive land exploitation, but also

* We are grateful to the Social Sciences and Humanities Research Council of Canada for research support (grant file no. 410-2009-159) and to the editor of this journal and two anonymous referees for many pertinent and useful comments. Bruce Campbell also contributed many excellent ideas and throughout has been a much appreciated supporter of this work. We also extend thanks to Philip Slavin for access to his photographs from the Northamptonshire Record Office, to Michael Fisher who prepared Map 1 for us, to the British Library for permission to reproduce Figures 3–5 , and to Catherine Glover for expert copyediting. Finally, we would also like to express gratitude to the staffs of all the record offices we visited throughout England during the summers of

2009 and 2010.

AgHR 63, II, pp. 187–220 187

188 agricultural history review

of alleged increasing unemployment and subsistence crisis.1 It would be useful to know how such imputed conditions for the period were matched by contemporary agricultural labour profiles, such as that of the famuli. In particular, it is argued here that the presence, or not, of women, children, and the elderly – as inferred here largely through grain payment levels and job descriptions – can not only illuminate the structure of such a workforce, but can also provide vital clues as to the health of the economy at any particular time.2

To set the famuli in context, they cannot be considered typical of all agricultural labour in England at this time, particularly the peasant farms (which we assume were more family- based), but they seemingly encompassed a full range of personnel from supervisors through to the most junior of workers. We estimate they comprised a total working population of 105,000 or so in England by the end of the thirteenth century (see Appendix A). Critically, this labour is very richly documented in manorial accounts,3 which, as part of monitoring agricultural operations as a whole on demesnes, tracked wages in kind and cash paid to each of the famuli workers,4 as well as indicating what that worker did, whether it was ploughing, carting, dairying, shepherding, or scaring away crows and rooks from newly seeded land (a particular duty of the young).

Although the famuli can only be considered the nucleus of the workforce needed for a typical demesne, since the customary working services of tenants and occasional ‘spot’ hiring of workers for particular tasks (especially weeding and, after the harvest, threshing) were also extensive, they comprised at least a third to a half of the demesne labour requirement.5

1 Any number of works can be cited for this view of English society around 1300, but a good summary of it (and competing visions for the period) can be found in John Hatcher and Mark Bailey, Modelling the Middle Ages: The history and theory of England’s eco- nomic development (2001), esp. ch. 2 (‘Population and resources’).

2 Following upon, say, Langdon and Masschaele’s contention that family income might be a better indi- cator of society’s well-being at the time than individual real wages: John Langdon and James Masschaele, ‘Com- mercial activity and population growth in medieval England’, Past & Present 190 (2006), pp. 35–81. To some extent the methodology suggested in this article has already been explored using royal works accounts: e.g., John Langdon, ‘Minimum wages and unemployment rates in medieval England: the case of Old Woodstock, Oxfordshire, 1256–1357’, in Ben Dodds and Christian

D. Liddy (eds), Commercial activity, markets and entre- preneurs in the Middle Ages: essays in honour of Richard Britnell (2011), pp. 25–44.

3 The latest (and very thorough) tally of mano- rial demesne accounts giving the sort of information used in this article puts the number at over 20,000, covering at least 2023 demesnes (a few of these are in Wales and Scotland, but the vast majority come from

England): Philip Slavin, ‘The sources for manorial and rural history’, in Joel Rosenthal (ed.), Understanding medieval primary sources: using historical sources to dis- cover medieval Europe (2012), pp. 131–48 (esp. pp. 132–6). Slavin estimates that there are an average of seven sur- viving accounts per documented demesne (p. 135), and that there are many demesnes that have exceptional runs over decades and even centuries (pp. 132–3).

4 There were also other perquisites often given to the famuli, such as daily portions of oats/peas pottage and celebratory ‘feasts’ at Christmas, Easter, and other times: see Appendix B.

5 Eona Karakacili provides a detailed example for Elton, Huntingdonshire, in 1323–4, where the famuli contribution was 43 per cent of the total labour needed for the demesne: ‘English labor productivity rates before the Black Death: A case study’, JEcH 64 (2004), pp. 24–60 (esp. p. 55). Christopher Thornton has also calculated that the proportional contribution offamuli labour was 42 per cent for the demesne at Rimpton, Somerset, around 1300: ‘The determinants of land pro- ductivity on the bishop of Winchester’s demesne of Rimpton, 1208 to 1403’, in Bruce M. S. Campbell and Mark Overton (eds), Land, labour and livestock: histori- cal studies in European agricultural productivity (1991), pp. 183–210 (esp. p. 205).

The famuli were particularly oriented towards soil preparation, especially ploughing, perhaps because it was felt that this early stage of crop production would be better served by a relatively stable workforce.6 As a result, more seasonally restricted activities like the harvest and haymaking do not appear strongly in the famuli documentation, although they were clearly expected to assist.7 Even with these exceptions, the range of work carried out by the famuli was nonetheless extensive enough across the arable and pastoral operations of demesnes to provide a useful labour profile, through which, with a carefully applied methodology, we can deduce much about its gender and age makeup, even if age in particular is very poorly revealed in any exact sense.

I

The two foundational studies on the English famuli are those of Michael Postan and David Farmer,8 and the famuli still remain the object of attention for other scholars looking for sets of consistently recorded labour.9 Both Postan and Farmer noted a key complication about the group in distinguishing between ‘service’ and ‘stipendiary’ famuli.10 The former worked for the relief of rents and/or labour services on lands that they held, while the latter worked for grain and cash wages. It seems probable, based upon Postan’s and Farmer’s views, that most of the famuli were originally of the service type but that gradually stipendiary famuli became more common.11 As Farmer observed, the economic rationale for this is not entirely clear, since service famuli seem to have been the far better option for lords in not requiring cash and grain outlays (see Appendix B), but famuli work performance might have improved under a wage regime.12 Indeed, it is important to note that both Postan and Farmer were examining demesnes from estates, principally those of the abbot of Glastonbury and the bishop of Winchester, where, by 1300, the proportion of service famuli was still significant. Demesnes in the rest ofthe country had by then swung mostly to using stipendiary famuli, so that – overall across England – these waged personnel comprised around 90 per cent of the 105,000 total famuli workers by c.1300 (Appendix A), a fact which makes this study particularly feasible.13

6 In part resonating with David Stone’s argument that hired labour was more productive on a per person basis than that supplied by tenant labour services: ‘The productivity of hired and customary labour: Evidence from Wisbech Barton in the fourteenth century’, EcHR 50 (1997), pp. 640–56.

7 As indicated by references to (probably young) people guarding working animals while the famuli went to the harvest (discussed below). For the famuli involvement in haymaking, see Stone, ‘Productivity’, p. 647n.

8 M. M. Postan, The famulus: the estate labourer in the XIIth and XIIIth centuries, (Ec.HR Supplement 2, 1954); David Farmer, ‘The famuli in the later Middle Ages’, in Richard Britnell and John Hatcher (eds), Progress and problems in medieval England: essays in honour of Edward Miller (1996), pp. 207–36.

9 E.g., Ian Rush, ‘The impact of commercialization in early fourteenth-century England: some evidence from the manors of Glastonbury Abbey’, AgHR 49 (2001), pp. 123–39.

10 The terminology is that coined by Farmer (‘Famuli’, p. 208); Postan was more vague about the distinction (e.g., Famulus, p. 4).

11 Postan, Famulus, p. 27; Farmer, ‘Famuli’, pp. 208–9.

12 Farmer, ‘Famuli’, p. 208; see also n. 6 above.

13 Service famuli are difficult to factor into the detailed statistical analysis, so no attempt was made to do so in this study, in effect treating them like tenants supplying labour services. Indeed, the only estate with similar proportions of service famuli working on its demesnes as we found for the bishopric of Winchester and Glastonbury Abbey was that of the Priory of Win- chester Cathedral, also in the same region. For the rest

190

agricultural history review

The accounts are, for the most part, also remarkably uniform country-wide in how they recorded the information about these stipendiary servants. In particular, the payment in kind made to a famulus/famula, usually called a ‘livery’ (liberatio), was largely recorded in terms of the number of weeks’ work needed to earn a quarter (8 bushels) of grain and/or legumes, which allows a close comparison from worker to worker. Most of the famuli, particularly the more established ones, were also given a cash payment as well. However, these payments in cash, worth about a quarter of the value of the grains that these workers received,14 are not as useful analytically as the grain payments, so we have treated them in the main text as incidental data, to be cited occasionally when relevant. Whether or not these combined remunerations of grain and cash plus other perquisites amounted to ‘living wages’ – for they were hardly generous – is, of course, an important question, but it is more complicated than at first appearance, and we have reserved a more detailed discussion of it for Appendix B.

Also, because it was not possible to do this in a consistent fashion, we did not take into account the type of ‘grains’ (which included legumes like peas and beans), that each worker received, as much as this would be useful in order to estimate, say, caloric equivalents received per worker. Many manorial accounts do indicate the type of grain given to each worker (and we have supplied that information in our examples when available), but most often the entire famuli were collectively given a ‘mixture’ (mixtura) of grains, ranging from wheat to peas or beans as recorded at the beginning of the section dealing with the famuli’s grain liveries, but without differentiating who got what beyond the amount of this ‘mixture’ each received. Even murkier were the cases where the grains making up the liveries were partly or wholly comprised of multure from manorial mills, where the types of grain were not indicated at all but simply expressed as quarters received ‘from the mill(s)’. In the same vein, we did not differentiate between the types of measurement for the grain (struck versus heaped bushels, for example), again because of the difficulties of doing this consistently across the sample. In short, we concentrated upon that most consistently and clearly supplied metric, the number of weeks that a famulus/famula was required to work to earn a quarter (regardless of what kind of grain/legume this was). This, for us, provided the most uncluttered source of information in deciphering status levels among these famuli, as well as, critically, revealing something of their gender and age composition.

Also, demesne accounts across the country seem unfailingly to have used a livery rate of 16 weeks or more per quarter as denoting a decidedly secondary pool of labour. Those within this less generous range of livery rates had – certainly on average – shorter periods of employment throughout the year; were characterized by a terminology of subordination (ancilla, garcio and so on); and, for the most part, performed a range of agricultural duties that have long been associated with neophytes and underlings. For the rest of this investigation, as an exercise in breaking down the constituents offamuli labour, particularly by gender and age, we are going

Note 13 continued

of the estates in our database – e.g., those of the earl of Lincoln, Westminster Abbey, Canterbury Cathedral Priory, etc. – stipendiary famuli were overwhelm- ingly predominant. For more comparisons of service (or ‘famuli in serjeanty’, as Farmer calls them) with

stipendiary famuli over geography and time, see David

L. Farmer, ‘Prices and wages’, in H. E. Hallam (ed.), The agrarian history of England and Wales, II, 1042–1350 (1988), p. 731.

14 See Appendix B under ‘Cash Stipends’.

to examine these demesne workers through a two-part division denoted by the 16 weeks per quarter livery and will characterize the parts as ‘first tier’ and ‘second-tier’ respectively in terms of pay and authority. First-tierfamuli (ploughmen, carters, shepherds and so on) working less than 16 weeks per quarter were mostly able-bodied adult males. Second-tier workers, having to work 16 weeks or more per quarter, had larger proportions of women but also, as we shall see, also included significant numbers of the young and the elderly, and possibly even the poor and disabled. Such employees generally supplemented or enhanced the activities of the first-tier workers, but, especially for young males, they might also be trainee labour hoping for eventual promotion to first-tier ranks.

Two important lacunae in the recorded liveries to famuli need to be emphasized. The first is that ages were never given for any of the famuli in the accounts, and so we do not have even occasional age-specific data to help guide our examination. Thus, the presence of child, adolescent, or elderly labour is inferred throughout by the grain/legume livery payment rate that a famulus or famula received and, to some extent, the task she or he performed. The second is that manorial servants in the early fourteenth century were seldom named, but recorded anonymously as ‘ploughman’, ‘carter’, ‘shepherd’, and so on. This is certainly a severe restriction in trying to establish things like family connections among the famuli and also, to some extent, the gender of the worker. One might also question whether the same person was involved in a job continuously through the period stipulated or whether two or more unnamed adults might have cooperated in fulfilling the specified duties, either serially or at the same time. Generally speaking, however, there is no evidence that this ‘job-sharing’ took place,15 and the restricted period for many of the jobs (often of only a few weeks) suggests strongly that only one person did it, although the person nominally in the position may have brought ‘helpers’ probably drawn from his or her family (see Appendix B). Finally, a lack of names makes it difficult to figure out how employment among the famuli worked out in a life-cycle sense. Did the young people we seemingly observe entering the ranks of the famuli do so in order to make a long-term career in demesne agriculture, or were they there for mainly short-term employment, among other things, to enhance family income?16 These are things that we can only speculate about here, but getting some sense of the shape of demesne labour, even for as a limited period as here, will be a good start.

15 When names are very occasionally supplied (as for male ‘dairymaids’: see n. 39 below) it indicates that the job was in fact held by a single person.

16 Evidence from the demesne of Houghall, Durham, in the late fourteenth and early fifteenth century, where the famuli were seemingly hired on six-monthly con- tracts, indicated a quick turnover of famuli person- nel, even at the first-tier level, which would suggest a more casual attitude on the part of those going into such employment rather than making it the focus of a lifetime career: Richard Britnell, ‘Employment

on a northern English Farm, 1370–1409’, paper deliv- ered to the 45th International Congress of Medieval Studies, University of Western Michigan, Kalamazoo, Michigan, 23 May 2010. We are grateful to have had permission from Professor Britnell before he passed away to cite his paper. A summary of the paper can be seen at:www.medievalists.net/2010/05/23/employment- on-a-northern-english-farm-1370-1409 (last accessed 5 Feb. 2015). For a relatively recent survey of medieval life-cycle issues, see Deborah Youngs, The life cycle in medieval Europe, c.1300–c.1500 (2006).

192

agricultural history review

map 1: Location of demesnes in account sample, c.1300.

II

Given the large number of surviving accounts, to make this preliminary examination of the labour profile within the famuli more manageable we decided to concentrate on a relatively narrow range of years around 1300, effectively encompassing the entire decades of the 1290s and 1300s. Since accounts normally ran from Michaelmas (29 September – the traditional end of the harvest) to Michaelmas of the following year, this meant examining accounts from 1289–90 to 1310–11, a total span of 22 account years. We further restricted ourselves to taking only one account per manor, normally that closest to the year 1300 (1299–1300 was the account year most preferred, if it survived). The end result was a sample of 434 accounts, and hence manors, found in 428 different communities. As Map 1 shows, the coverage of the sampled manors across the country is uneven, being heavily skewed to the south and east of the country with notably ‘empty’ areas such as the forest area of the Weald south of London, the extreme South West (Devon and Cornwall), and the northern and western areas of the country generally, which largely reflects the regional survival of manorial accounts during this particular time period.

Labourers

No. of

1800

1600

1400

1200

1000

800

600

400

200

0

– 1537
	663
	388		384
– 143			148
118 0 0 0 0 1 6				79	99 18 22 18	2 16 3 2 7	49 7 3 2 0 0 0 1 10 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33+

Grain Payments (Weeks/Quarter)

figure 1: Grain Payment Rates for Demesne Labourers

Total Sample N=3748

Altogether we gathered data for 4581 stipendiary famuli from the 434 accounts, which were entered into a spreadsheet. Of these, the grain livery rate in number of weeks required to earn a quarter of grain was directly stated in the account or could be calculated – by dividing the number of weeks worked by the number of quarters paid – for 3748 (81.8 per cent of the 4581 total) of these workers, while the remaining 833 (18.2 per cent) only recorded the amount of grain given to the worker without specifying the time required to earn it. Figure 1 consequently shows the distribution of livery rates for the 3748 workers for which the more specific information is known, with the x-axis showing the number of weeks afamulus/famula worked to earn a quarter of grain, the better paid being to the left of the histogram and the poorer paid to the right, while the y-axis indicates the number of workers at each particular payment rate (this and other breakdowns ofthe data are presented in abbreviated tabular form in Appendix C). The distribution for the ‘first-tier’ workers is clear enough, with a very notable peak at 12 weeks required per quarter for 1537 of them (or 41.0 per cent of the total 3748). There was a wide variation around this mode value for first-tier workers, ranging from the single case of only five weeks required per quarter for a ‘seeder’ at Ickham, Kent, in 1294–95,17 to 18 cases at 15 weeks per quarter, which seems to have existed as a sort of transition zone between the first-tier and second-tier workforces. There were also notable concentrations at the eight and

17 Canterbury Cathedral Archives (hereafter CCA) DCc Ickham 12. This rate is not implausible, since consider- able skill was needed to ensure a uniform spread of seed over ploughed soil: e.g., Christopher Dyer, ‘Documentary evidence’, in Grenville Astill and Annie Grant (eds), The countryside of medieval England (1988), pp. 12–35 (esp. pp. 26–7); John Langdon, ‘Agricultural equipment’, in ibid., pp. 86–107 (esp. p. 99).

194

agricultural history review

ten weeks per quarter levels (10.4 and 17.7 per cent of the total 3748 respectively), which relate to traditional rates on particular estates.18

In Figure 1, the start of the ‘second-tier’ ranks is signalled by the significant number of cases at 16 weeks per quarter (384 or 10.2 per cent of the 3748 sample). When combined with the long tail of even lower rates (that is, the 164 cases from 17 to 32 and more weeks’ work required per quarter), the total second-tier personnel in the sample comes to 548, or 14.6 per cent of the total 3748, indicating that roughly one in seven workers was of this station. It is probable that this proportion is an underestimate, since the additional 833famuli for whom we could not ascertain the number of weeks per quarter probably had an even greater percentage in the second-tier ranks (see, for example, the analysis of bird-scarers below). On the other hand, second-tier personnel in the sample tended to work less often, on average only 30.5 weeks per year compared to 44.1 weeks per year for their ‘first-tier’ colleagues. As a result, the 3200 first-tier workers among the 3748 total for whom grain payment rates were known were employed for a total of 141,271.8 weeks (89.4 per cent) compared to 16,702.6 weeks (10.6 per cent) for the 548 persons in the second-tier category. The difference in the payment of grains for the two groups was even more pronounced, with first-tier workers receiving 13,363.5 quarters (or 93.6 per cent) compared to 920.5 quarters (or 6.4 per cent) given to the second-tier group. There is a marked regional variation in the proportion of first- to second-tier workers among the famuli, as shown in Table 1 in order of the amount of second-tier labour present.19 The North stands out as having the highest level of second-tier personnel across the board, from 20 per cent of personnel to over ten per cent of grains received, over double that, say, of the region with the least amount of such subsidiary labour, East Anglia, with the other regions falling in between. As Table 1 also shows, there seems to be an inverse relationship between population density in a region and its use of second-tier labour, perhaps implying that the North suffered labour shortage compared to, especially, population-rich East Anglia.20 Part of it, however, may also have been more managerial in nature, particularly as evident on the bursar’s manors of Durham Cathedral Priory, which seemingly had a more developed practice of recruiting and training new famuli (see below). On the other hand, the Thames Basin region had both relatively high population density and a relatively high use of second-tier labour.

18 For example, a rate of eight weeks per quarter (and sometimes better) seems to have been the case for ploughmen, carters, and the like on the Kentish manors of Canterbury Cathedral Priory, while ten weeks per quarter was common for such workers on many West- minster Abbey manors.

19 The proportions of second-tier people on ecclesi- astic versus lay estates were also determined at 15.6 and

12.4 per cent respectively, but t-testing indicated the dif- ferences were not (quite) statistically significant. Also, the uneven geographical distribution of data noted from Map 1 above did not seem to make a much difference to the proportion of second-tier workers for England as a whole. Using Campbell’s recent assessment of county populations in 1290 (Bruce M. S. Campbell, ‘Bench- marking medieval economic development: England,

Wales, Scotland, and Ireland, c.1290’, EcHR 61 (2008), pp. 896–948 (esp. Table 14 [p. 926]) and weighting each of the regional proportions of second-tier labour in Table 1 (in this article) according to the population for that region, results in an overall country figure of 14.5 per cent for the second-tier element in the famuli in terms of personnel numbers compared to the 14.6 per cent currently in Table 1. Similarly, the weighted figures in terms of weeks worked and for grain received are 10.8 per cent and 6.7 per cent respectively, again close to the 10.6 and 6.4 per cent figures currently in Table 1.

20 We are grateful to an anonymous referee for sug- gesting the possible connection of second-tier labour to population density. The population density figures in Table 1 were again calculated using data from Camp- bell, ‘Benchmarking’, Table 14 (p. 926).

table 1: Regional proportions of ‘first-tier’ versus ‘second-tier’famuli

		% personnel		% weeks worked		% grain received		Population
Region	No. of famuli	first-tier second-tier						Density (Persons/mi2)
				first-tier second-tier		first-tier second-tier
North	325	80.0	20.0	83.7	16.3	88.6	11.4	52.1
Thames Basin	1269	83.3	16.7	88.7	11.3	93.4	6.6	103.2
South and South West	822	85.5	14.5	89.8	10.2	93.8	6.2	72.4
Midlands	723	87.1	12.9	90.2	9.8	94.2	5.8	86.7
East Anglia	609	90.3	9.7	92.8	7.2	95.4	4.6	141.8
Total Sample	3748	85.4	14.6	89.4	10.6	93.6	6.4	79.1

Sources: Manorial Account Database for years 1289–90 to 1310–11.

Note: The counties within each region (excluding those counties with no data) are as follows, in order of the proportion of second-tierfamuli for each region: 1) The North (Cumb., Durham and Yorks.); 2) The Thames Basin (Beds., Berks., Bucks., Essex, Herts., Kent, Middx., Oxon. and Surrey); 3) The South and South West (Dorset, Devon, Hants., Somerset, Sussex, Wilts.); 4) The Midlands (Ches., Glos., Leics., Lincs., Northants., Notts., Rutland, Salop., Staffs., Warks. and Worcs.); 5) East Anglia (Cambs., Hunts., Norfolk, Suffolk).

It was, incidentally, also the region where famuli were most generously paid overall, with a median grain payment rate of only ten weeks work required per quarter of grain compared to a median of 12 weeks per quarter of grain for the other four regions.21 This phenomenon, coupled with a relatively high second-tier labour element, is probably a result of the Thames Basin being the most economically active area in the country, centred around London, giving both higher rewards to first-tier agricultural workers and greater opportunity for its second-tier ones.

(a) First-tier Workers

Here we have a quick summary of first-tier workers and the volatile first- versus second-tier split that could occur within various worker categories. Figure 2 shows this for the eight most numerous types of workers in the sample. As the figure demonstrates, the demesne workforce was clearly centred around the three most frequently recorded of the ‘first-tier’ famuli – ploughmen (carucarii, famuli carucarii, tentores, or fugatores), carters (carectarii), and shepherds (bercarii). In all three of these categories, the proportion of personnel paid at second-tier rates comprised five per cent or less (see Appendix C, part 2).

Ploughmen were predominant in number at 1423 (or 38.0 per cent) of the 3748famuli with specified grain livery rates.22 They themselves were usually divided into ‘holders’ (tentores),

21 As evident in appendix C, part 1, where the rep- resentation of the higher rates of payment (especially at the 5–7 and 8–9 weeks per quarter levels) is much greater in the Thames Basin than elsewhere. See also the generous liveries of two estates prominent in the region, Canterbury Cathedral Priory and Westminster Abbey: n. 18 above.

22 This should be considered as a minimum, since some ambiguous terms were not included among the 1423 ‘ploughmen’, such as bovarius (literally ‘ox-herd’, but indicating a ploughman – there are 116 cases of them in the sample) or just famulus (also in many contexts probably a ploughman – 77 of them in the sample).

196

agricultural history review

1600

1400

1200

No. of Laborers

1000

800

600

400

200

0

99%

96%

95%

83%

92%

73%

14%

1%

34% 17%

4%

5%

27%

8%

86%

66%

Dairymaids

Shepherds

(n = 346)

Carters

(n = 427)

Ploughmen

(n = 1423)

Cowherds

(n = 133)

Supervisors

(n = 166)

Garciones

(n = 153)

Harrowers

(n = 61)

(n = 192)

Type of Labourer

Second-Tier First-Tier

figure 2: Composition of labour in eight largest famuli roles

those who held the plough-handles and managed the tricky job of guiding the plough at a constant depth through the earth, and ‘drivers’ (fugatores), those who drove on the plough- animals, usually oxen, with a goad or whip, as shown in the famous Luttrell Psalter ploughing illustration (Figure 3). The holder was the more senior and experienced of the two, but this was generally not reflected in a greater amount of grain received, since both holder and driver normally received the same livery, but in a slightly higher cash payment given to the holder.23 However, when new recruits entered the ploughmen’s ranks, it was generally as a fugator first, as shown in a 1299–1300 Bewley, Durham, account, where, among a very large contingent of plough-people, there were also ten ‘pages driving the ploughs’ on the manor, who seemed to have been trainee labour coming into the ranks of the fugatores (see also the discussion of ‘pages’ below). Occasionally, if there were numerous ploughs and ploughmen on a manor, a ‘master ploughman’ would be designated.24

Carters (427, or 11.4 per cent, of the 3748 sample) were less hierarchical. Generally there was only one on a manor, but two or more might be found on larger enterprises, say on manors with over 300 sown acres. Occasionally ‘second’ carters were named and might be included in the

23 As one example among many, all ploughmen on the Westminster Abbey manor of Knowle, Warks., both tentores and fugatores, worked 12 weeks for a quarter of grain, but the former received 5s. cash for the year (1298–99) while the latter only got 4s.: Westminster Abbey Muniments (hereafter WAM), 27695.

24 This was particularly the case on the Kentish

manors of Canterbury Cathedral Priory, where ‘master’ ploughmen were found at Barksore in 1298–99 (CCA, DCc Barksore 14), Copton (in Preston) in 1291–92 (Copton 1), Elverton (in Stone, near Faversham) in 1297–8 (Elverton 9), etc., although they were not paid anything more in grain and cash than other ploughmen.

figure 3: Luttrell Psalter ploughing scene

(© The British Library Board Add. MS 42130, fo. 170)

second-tier ranks,25 but carters were almost always first-tier members in the famuli workforce. In contrast, shepherds (bercarii: 346, or 9.2 per cent, of the 3748 total) were arranged much more by status and experience, as seen in the Peterborough Abbey manor of (North) Collingham, Nottinghamshire, in 1300–01, with a ‘shepherd’, ‘second shepherd’, ‘third shepherd’, and a ‘boy shepherd’ (garcio bercarius) being recorded. The first three were all given a ‘full livery’, which, for this manor, required each of them to work ten weeks to receive a quarter of mixed grains (mostly rye plus grains received from the manorial windmill), while the garcio bercarius was given a ‘half livery’ requiring 20 weeks work per quarter.26 Indeed, as we shall see again below in a fuller assessment of the total ‘sheep carer’ population, many working with sheep were not labelled specifically as ‘shepherds’ (that is, the bercarii represented in Figure 2), but rather as ‘keepers’ (custodes), being responsible for particular segments of the manorial flocks, such as the ewes, ‘hoggs’ (hoggastri; young castrated males), lambs, and sometimes even rams.

The final group from Figure 2 indisputedly in the first-tier category are those we have categorized as ‘supervisors’, which were called variously in the accounts, in order of their number as stipendiary famuli, reap-reeves (messores), haywards (haywardi), serjeants (servientes), reeves (prepositi), beadles (bedelli), and bailiffs (ballivi).27 They were usually recruited from the ranks of experienced landholders,28 and indeed, more than any other group in the first-tier ranks, they were often servicefamuli or paid cash only, especially reeves and bailiffs. Of the 166 supervisors who were stipendiary famuli and thus included in Figure 2, the median payment required them to work 12 weeks per quarter of grain, but there were very significant levels of higher payments, especially at the eight weeks per quarter level (51 or 30.7 per cent), and they could sometimes be given superior grains.29 Yet, curiously, 13 of these supervisory personnel (or 7.8 per cent of

25 As at Milton Hall (in Prittlewell), Essex, where a ‘second carter’ worked for ten weeks at a rate of one quarter per 16 weeks’ work: CCA, DCc Middleton 16, m. 1d.

26 Northamptonshire RO (hereafter NRO), F(M) Charter/2388, ms. 18r and 18d.

27 The surprisingly low numbers of reeves and bail- iffs in particular are because most existed as service famuli in the accounts.

28 E.g., David Stone, Decision-making in medieval agriculture (2005), pp. 13–14.

29 As at Milton Hall (n. 25 above), where the ‘serjeant’ was given wheat at a rate of a quarter per eight weeks’ work, in comparison to the rye given to the other first-tier workers at rates of a quarter per ten weeks (for a carter and a shepherd) or per 12 weeks (four ploughmen).

198

agricultural history review

the total 166) in Figure 2 were paid at Ôsecond-tierÕ rates requiring 16 weeks of work or more per quarter (and for whom there was apparently no additional compensation such as relief of rents or labour services). Ten of these 13 were designated as messores, generally associated with overseeing the harvest, but often called on for other duties.30 Occasionally these lowly paid messores were lumped in with other patently second-tier workers, as at Ketton, Durham, in 1299Ð1300, when the messor was grouped with two pages, one swineherd, and the dairymaid, each getting one quarter per 16 weeks.31 As we shall discuss later, one explanation might be that these Ôsecond-tierÕ messores were elderly people whose physical capabilities no longer commanded a first-tier livery, but whose age provided sufficient authority for supervisory work.

(b) Women: first- or second-tier workers?

We move now to those workers who gravitate more to the second-tier side of the spectrum. Here the picture becomes more complicated and gender issues start to play a considerable role. The two groups in Figure 2 most involved here are the ÔcowherdsÕ (vaccarii) and the ÔdairymaidsÕ (dayae or daiae). The term vaccarius for the cowherd seems to stamp the position as one for males, and fewer of them Ð 17.3 per cent Ð were at the second-tier level compared to 26.6 per cent for the dairymaids. As might be expected, though, the position was connected to that of the dairymaid and at times was clearly interchangeable with it and may have been seasonal, so that, on the manor of (Old) Bolingbroke, Lincolnshire, in 1295Ð96, the dairymaid there was the vaccarius in winter.32 But cowherds could double as other types of (apparently male) workers, as at Sedgebrook, Lincolnshire, in 1295Ð96, when the cowherd also drove ploughs on occasion.33 This may simply underline that cowherds tended to be of a lesser rank among the famuli, but in a trend that foreshadowed later times, some were also becoming entrepreneurial, as in the case of the cowherd at Little Chart, Kent, who was given one seam (the Kentish version of a quarter) per eight weeks for 31 weeks during 1301Ð02 for a total of three seams and seven bushels, clearly a Ôfirst-tierÕ rate, but only one seam and two and a half bushels for the remaining 21 weeks of the year (a rate of just over 16 weeksÕ work required to earn a quarter) Ôbecause he had the dairy at farmÕ (that is, the herd was leased to him).34

The interchangeability of cowherds and dairymaids inevitably throws up the question: what (or who) was a ÔdairymaidÕ? Since the Latin daya or daia is feminine and milking was clearly associated with women,35 it might be easy Ð somewhat reflexively Ð to consider them all as

30 E.g., Mark Page (ed.), The pipe roll of the bishopric of Winchester, 1301–2 (Hampshire Record Ser., 14, 1996), p. 375.

31 In liberatione duorum pagiorum unius porcarii unius messoris unius daye dictum tempus [i.e., one year] quarterio dato per xvj ebdomadas, xvj quarteria j rasar- ium: Durham University Library, Dean and Chapter of Durham (henceforward DUL, DCD) Enrolled Manors, 1299Ð1303, m. 3r.

32 … in stipendiis … j daie qui est vaccarius in yeme: TNA, DL 29/1/1, m. 8r.

33 Et de iij quarteriis ij bussellis in liberatione j qui fuit vaccarius in yemale & fugans carucam per vices:

TNA, DL 29/1/1, m. 1d.

34 … quia habuit daeriam ad firmam …: CCA, DCc, Little Chart 6. Farmer noted this trend of leasing the demesne herd as becoming common in the later four- teenth century: ÔThe FamuliÕ, p. 224.

35 E.g., the women milking a cow with calf in MS Bodley 764 (as shown in English rural life in the Middle Ages (Bodleian Picture Book 14, 1965), plate 5a), and the women in the sheep-milking scene in the Luttrell Psalter: British Library Add. MS. 42130, fo. 163 (shown, for example, in Janet Backhouse, Medieval rural life in the Luttrell Psalter (2000), p. 30).

female. The distribution of grain payments for dairymaids, as shown in Appendix C, Part 2, certainly indicates a less generous remuneration for them as against, say, the profile for all workers in Figure 1. This might well support the findings evident in so many other forms of remuneration between the sexes that medieval women were paid less than male contempo- raries when doing similar types of work.36 However, muddying this conclusion is the fact that some at least of the ÔdairymaidsÕ were apparently males.37 Examples include Ôthe man (homo) making the dairy [work] and the pottage for the famuli and winnowing all the corn of the manorÕ at Kings Langley, Hertfordshire, in 1305Ð06 for the year at a livery of one quarter of grain per 12 weeks, or the man (again homo) making Ôthe office of vaccarius and dayaÕ at Laleham, Middlesex, in 1304Ð5, this time at one quarter for every nine weeks.38 To these can be added occasional references to ÔdairymaidsÕ sporting masculine names,39 which should make us wary of assuming that all such designated dairy personnel were necessarily female.

There are plenty of other references, however, that indicate that the ÔdairymaidÕ was normally female. In a 1300Ð01 account for Castor, Northamptonshire, it stated that a daya was given a livery at a rate of one quarter per 12 weeks except for four weeks in the harvest and two weeks in quibus nulla erat daya (Ôin which no one was the dairymaidÕ), the nulla in Latin indicating that the dairymaid was indeed a female here (unless the scribe was more interested in making the Latin agree than in reflecting gender reality).40 Similarly, if the person was a male some scribes felt they had to indicate this, as in the 1301Ð02 Bishops Sutton (Hampshire) account reference to a livery of one quarter per ten weeks given to Ôone man who is in place of the dairymaidÕ,41 suggesting the position was normally one for a female. The agricultural treatises of the time also seem to have leaned toward the dairymaid as being female. The Seneschaucy, seemingly written between 1260 and 1276, used the pronoun ÔsheÕ (ele) throughout when considering the ÔofficeÕ of dairymaid.42 The anonymous Husbandry, written closer to 1300, took a more cautious line, indicating that the ÔdairymaidÕ could be a man but also in the process referring mostly to the dairy-person as a female:

And you ought to have in every place where there is a dairy some person in charge [ou daerye est une daye], be it a man or a woman. And if it were a man he ought to do the same things a dairymaid would do. And, because of the benefits which he has from milk he ought to take one quarter of corn every sixteen weeks where other servants have one quarter for every twelve weeks.

36 Sandy Bardsley, ÔWomenÕs work reconsid- ered: Gender and wage differentiation in late medi- eval EnglandÕ, Past & Present 165 (1999), pp. 3Ð29 (esp. pp. 11Ð12); Langdon, ÔMinimum wagesÕ, esp. pp. 28Ð36.

37 Farmer also made this point: ÔFamuliÕ, p. 224.

38 TNA, SC 6/866/16, m. 1d; WAM, 27114, m. 1d. It is assumed that homo means ÔmanÕ here rather than, say, the ambivalently gendered Ôhuman beingÕ, which could include a woman or girl. Indeed, using homo to indicate a female in a job that was largely considered female anyway would seem an unnecessary ambiguity.

39 Thus the references to Richard Ôle DayeÕ at Ches- terton, Essex, in 1301Ð2 (TNA, SC 6/837/24, m. 1r) and

Nicholas Daye at Popenhoe, Norfolk, in 1291Ð2 (TNA, SC 6/942/13, m. 1d). Similarly there is a reference to a reasonably generous grain livery of one quarter per 12 weeks given to Ôle deyÕ (rather than Ôla deyÕ) in the Fornham St. Martin, Suffolk, account of 1300Ð1: Suffolk RO, Bury St. Edmunds branch, E3/15.9/2.11, m. 1d.

40 NRO, F(M) Charter/2388, m. 5d.

41 Page (ed.), Pipe roll of … 1301–2, p. 308; see also

p. 307.

42 Seneschaucy, cc. 66Ð69, in Dorothea Oschinsky, Walter of Henley and other treatises on estate manage- ment and accounting (1971), pp. 287Ð8. For the dating of this treatise, see ibid., p. 89.

200

agricultural history review

And the dairymaid ought [E ele deyt] to winnow all the corn, and half of her pay shall be

for paying the woman [femme] who will help her.43

The last sentence in this excerpt, and a slightly later reference to the dairymaid (in the feminine) also being required to look after Ôsmall stockÕ, including poultry and eggs,44 as well as the statement that her wages should be shared by her helper (for more on these ÔhelpersÕ, see Appendix B), suggest a lower individual pay and status for females in the ÔdairymaidÕ position. From this, it might follow that a larger proportion of male ÔdairymaidsÕ would occupy the higher grain payment group (that is, getting more than one quarter per 16 weeks despite what the Husbandry advised), while women would more often be found in the lower-paid group getting one quarter per 16 weeks or less. Consequently, when only those people in the sample who were undisputedly women Ð labelled as mulieres (probably adult women), ancillae (that is, maidservants, probably young women or adolescent girls45) or feminae Ð are considered, they mostly fell within the Ôsecond-tierÕ group. Although the sample is small Ð at 46 individuals Ð 40 of them (or 87.0 per cent) had to work 16 weeks or more for a quarter of grain (see Appendix C, Part 2). The descriptions of what work these particular women performed indicate that a good deal of it centred around the manorial complex of buildings, the curia as it was often called, so that 14 of these women (or 29.8 per cent) were described as ancilla domus, mulier custodiens domum curiae, or something similar.46 They also did jobs like making the oat pottage for the famuli, winnowing grain, milking ewes, and drying malt.47 At Caistor cum Markshall, Norfolk, in 1299Ð1300 (or possibly 1300Ð01) one of them seems to have probably started out as an ancilla for 25 weeks before being promoted to a daya for another 23 weeks.48

However, despite the possibility that many women might have been in the better-paid group of dayae, female dairymaids were clearly in a more liminal position than more

43 Husbandry, c. 13, in Oschinsky, Walter of Henley, p. 425. For the dating of the treatise, see ibid., pp. 200Ð1.

44 Ibid., c. 16 (p. 425). For an example of a Ôdairy- maidÕ also being expected to take care of poultry from our account sample, see Page (ed.), Pipe roll of … 1301–2, p. 257 (Bishops Waltham, Hants.).

45 We do not go as far as Susan Mosher Stuard in considering ancillae as some form of slave labour (ÔAncillary evidence for the decline of medieval slaveryÕ, Past & Present 149 (1995), pp. 3Ð28), since they often seem to have been considered the equal of, say, muli- eres in such situations. In this, our position follows that of Jean-Pierre Devroey, ÔMen and women in early medieval serfdom: The ninth-century north Frank- ish evidenceÕ, Past and Present 166 (2000), pp. 3Ð30 (esp. pp. 29Ð30), in seeing a fundamental legal equality between these (girls?) and other men and women in peasant society, although their generally low status is abundantly clear.

46 For example, taking two examples from the extremes of payments to these women, an ancilla domus at Chaddington, Bucks., in 1302Ð3, worked 12

weeks for each quarter of grain she received (Merton College Library, Oxford [hereafter MCL] 5537), while a mulier keeping the Ôhouse(s) of the court and making the pottage of the famuli Õ at Hurcot, Somerset, in 1300Ð1 only received five bushels for what was claimed to be an entire yearÕs work, a rate requiring over 82 weeksÕ work per quarter of grain: TNA, SC 6/1090/6, m. 3d.

47 As, respectively, at Upton, Northants., in 1300Ð1 (NRO, F(M) Charter/2388, m. 22d); Thorpe (in Peter- borough), Northants., again in 1300Ð1 (ibid., m. 22r); see also similar cases at Pittington, Durham, in 1299Ð 1300 (DUL, DCD Enrolled Manors, 1299Ð1300, m. 2r); and Therfield, Herts., in 1306Ð7 (TNA, SC 6/872/17, m. 4r).

48 In liberatione unius ancillae a festo sancti Michaelis [29 Sept.] usque festum Annunciationis beatae Mariae [25 Mar.] per xxv septimanas, iiij busselli. In libera- tione unius dayae a festo Annunciationis beatae Mariae usque festum sancti Michaelis praeter iiij septimanas in autumpno per xxiij septimanas, j quarterium iij bussel- lis dimidium: TNA, SC 6/932/26, m. 1d.

well-established, continuously employed famuli like ploughmen and carters, particularly if some of the dairymaids indicated in Appendix C, Part 2, and especially the better-paid ones, were in fact males. In short, the gender makeup of so-called ÔdairymaidsÕ seems to have become increasingly fluid, particularly with the leasing of demesne herds, although the term daya or some form of it was still associated with women even in ChaucerÕs day with his reference in the NunÕs PriestÕs Tale to a poor widow who made her living as a ÔdeyeÕ.49 Dairying as a consequence seems alive with the sort of gender complications that Judith Bennett highlighted in her study on late medieval and early modern brewing.50

(c) Second-tier workers: the young

We now move onto what appears to have been child or adolescent labour in the sample. The most obviously young, or at least ÔtraineeÕ, were those named as ÔpagesÕ (either pagii or pagetti; but most often abbreviated to pag’ in the documents). Household accounts suggest that they were younger than the garciones discussed below, perhaps, as C. M. Woolgar has suggested, being Ôprobably pre-adolescentÕ.51 In our sample they were found infrequently (in only 21 cases), mostly on northern manors and in contexts that indicated they were very junior; when the rate of grain payment was either given directly on the document or could be calculated, it was almost invariably at one quarter per 16 weeks or less.52 In some cases, it was obvious that these pages were part of a graduated training process, as on the large manor of Bewley, Durham, in 1299Ð1300, where ten Ôpages driving the ploughsÕ were recorded as working for 29 weeks at a stated rate of one quarter wheat for 24 weeks each; they were at the tail end of a ploughing hierarchy that involved 20 full-time (that is, for the entire year) and four part-time ploughmen, paid at an equivalent rate of one quarter per 12 weeks of (mostly) wheat each.53 Similarly a page driving the plough in the harvest was recorded for Little Langton (between Great Langton and Thrintoft), Yorkshire, in 1304, at an equivalent rate of one quarter (of rye) per 16 weeks.54 Pages helping shepherds were found intermittently in the sample, as at Pittington, Durham, in 1299Ð1300, where two pages and a woman (mulier) helped a shepherd, especially at lambing time,55 while again at Little Langton in 1304 another page kept calves.56

49 Larry D. Benson (ed.), The Riverside Chaucer (third edn, 2008), p. 253, l. 2846.

50 Judith M. Bennett, Ale, beer, and brewsters in England: women’s work in a changing world, 1300–1600 (1999).

51 C. M. Woolgar, The great household in late medi- eval England (1999), p. 40.

52 The one case of a ÔpageÕ being paid more than one quarter of grain per 16 weeks of work was at Stalling- borough, Suffolk, in 1307, where a page was given six bushels of wheat for eight weeksÕ work for a variety of chores Ôin the time of lambing, weaning, and carrying milk [presumably from the ewes]Õ: Suffolk RO, Ipswich Branch MA53 359/354 (iii), m. 1d. This is probably erro- neous, since this rate, only requiring 10.67 weeksÕ work per quarter, was the best among the famuli on the manor. What seems most probable is that the eight

weeks were for the lambing season only (normally in the range of a month to 14 weeks, traditionally starting from the Purification of Mary, 2 February), while the supervision of the weaning of lambs and carrying of milk added extra weeks not recorded.

53 DUL, DCD Enrolled Manors, 1299Ð1303, m. 1r. The demesne sown acreage was probably around 635 (as estimated from the number of quarters sown).

54 North Yorkshire Record Office (hereafter NYRO), ZJX 3/2/12, m. 1d.

55 Cuidam pagio adiuvanti hau [?; high?] bercario per sexdecim ebdomadas, alio pagio per mensem tempore agnelationis & mulieri querenti lac ad agnos, j quarte- rium ij rasaria: DUL, DCD Enrolled Accounts, 1299Ð 1303, m. 2r.

56 Et in liberatione unius pagii custodientis vitulos, j estricha: NYRO, ZJX 3/2/12, m. 1d.

202

agricultural history review

However, a much larger group of possibly young workers in the sample were those styled as garciones, comprising the sixth largest grouping in Figure 2. Household accounts suggest they were adolescents and sometimes rowdy ones,57 but otherwise the type of person represented by the term garcio has been very hard to pin down. It might well signify someone young but it could just as easily represent a – most probably male58 – servant of any age. Harold Fox, who has supplied the most detailed discussion to date of the term garcio within a manorial context, was categorical in not confining it to a particular age group: ‘Garcio, then, is not specifically the terminology of youth … ; suffice it to say here that we are dealing with a term which etymologically implies low status and menial work and was used in this sense before also coming to designate a youth,.59

The distribution of livery rates for garciones shown in Appendix C, Part 2, certainly does not contradict Fox,s definition, where the ‘low status, of these workers is amply confirmed, as 131 (or 85.6 per cent) of the total 153 garciones had to work 16 weeks or more for a quarter of grain (the mode here was very strongly at one quarter per 16 weeks worked, where 89 of the 153 garciones – or 58.2 per cent – received exactly this grain livery rate). Certainly the potential age range of garciones seems to have been extensive. Some were almost certainly adults, as in the case of the garcio, who, by order of the bailiff, supervised the threshing and winnowing, at a first–tier livery of eight weeks per quarter, half of wheat and half of barley, on the bishopric of Winchester manor of East Knoyle (Wiltshire) in 1301-02.60 Many, on the other hand, were undoubtedly young or still subordinate to parental authority, as at Westerham, Kent, in 1296-97, where a garcio was paid at a rate of one quarter per 20 weeks for guarding the Abbot of Westminster,s sheep along with the sheep of his father.61

Our sense, though – following Fox as seeing context as key in deciding how the term garcio should be interpreted62 – is that most garciones within the context of the famuli were probably young and some very young. This is perhaps best seen through those designated as bird–scarers keeping crows, rooks, and other birds from newly sown crops, a traditionally neophyte activity seemingly performed with sling–shots, as shown in Figure 4. Altogether, bird–scarers were recorded in 26 of the 434 accounts (6.0 per cent) and were found almost solely in southern parts of the country, where concern about maximizing arable production was seemingly strongest.63

57 Woolgar, Great household, pp. 39-40. Woolgar translates garcio as ‘groom,.

58 Female garciones were never directly indicated in this account sample, so we have assumed garciones were always males, following Fox who clearly felt this:

H. S. A. Fox, ‘Exploitation of the landless by lords and tenants in early medieval England,, in Zvi Razi and Richard Smith (eds), Medieval society and the manor court (1996), pp. 518-68. Occasionally, however, gar- ciones might do things more associated with female workers, such as making oat pottage and doing house – hold duties around the manorial range of buildings (e.g., see the 1308 Broadwell, Oxfordshire case below), suggesting the possibility that a few garciones might have been female.

59 Fox, ‘Exploitation,, p. 521.

60 Page (ed.), Pipe roll … of 1301–2, pp. 51-2. Page translates the Latin garcio as ‘attendant, (p. 369), reflect– ing to some extent the confusion over the term; see also

n. 58 above.

61 In liberatione j garcionis custodientis bidentes domini una cum bidentis patris sui per annum, ij quar- teria iiij busselli dimidium, per xx septimanas quarte- rium: WAM, 26389, m. 2d.

62 As he amply demonstrates when coming to a somewhat different interpretation of garcio as a landless male of at least 12 years old and above: ‘Exploitation,, esp. pp. 520-1.

63 From east to west across the south of England, the counties recording bird–scarers (number of manors in brackets) were Norfolk (2), Suffolk (2), Essex (7), Kent (2), Herts. (2), Middx (1), Surrey (1), Sussex (1),

figure 4: Luttrell Psalter scaring crows and harrowing scene

(© The British Library Board Add. MS 42130, fo. 171)

Of these, 14 were designated as garciones. Where the rate of grain payment was indicated – in eight of these 14 cases – six of these garcio bird-scarers had to work 16 weeks for their quarter of grain, one for 18 weeks, and one for 32 weeks. Three of the remaining 12 cases of bird-scaring, but where the person was not styled as a garcio, also had to work 16 weeks for their quarter of grain. How old might these bird-scarers have been? Jane Humphries, in her recent book on child labour during the Industrial Revolution, gives several examples of bird-scaring (of crows or rooks usually) as the first job that young people were given in an agricultural setting. Of the 22 or more instances of crow-scarers Humphries found in her sample of diaries from the period,64 the ages of three of them when they started crow-scaring are recorded in her text as ‘nine’, ‘not yet six’, and ‘from age six’.65 A fourth and particularly illuminating example was that of William Arnold, born in 1860, whose first job was ‘scaring crows from newly sown fields in late February and early March’, before he went on to guard 100 sheep, lead the first horse of the wagon and mind 40 pigs during the acorn season, all before he went into the boot trade ‘aged just over seven’.66

The account material also gives this sense of bird-scaring as a starter position. At Kennett, Cambridgeshire, in 1299–1300, a garcio was hired to guard the manorial curia (or range of buildings for the demesne) against rooks ‘lest they should nest within’ for what looks to be a modest half-quarter (four bushels) of grain for an unspecified period of time.67 A similarly poorly rewarded task for an unspecified period was recorded for ‘a keeper of the wheat in winter because of wild geese’, for which the person involved was only given two bushels, at Little Humber, Yorkshire, in 1296–97.68 Presumably if the ‘scarer’ was good at it, it could be

Note 63 continued

Berks. (1), Hants. (4), Dorset (1), and Somerset (1). The single outlier outside this southerly band was Little Humber, Yorks. (mentioned below). As indicated, bird- scaring was most commonly found in Essex, where, in the sample, seven of the 18 manors (or 38.9 per cent) for the county recorded some degree of the practice.

64 Jane Humphries, Childhood and child labour in the

British Industrial Revolution (2010), p. 211.

65 Ibid., pp. 174, 188, 230.

66 Ibid., p. 219.

67 In liberatione j garcionis custodientis curiam pro fruibus ne intus nidarent, dimidium quarterium: TNA, SC 6/768/20, m. 1d.

68 In liberatione j custodis frumenti in yeme propter aucas sauvagnias, ij busselli: TNA, SC 6/1079/15, m. 4d.

204

agricultural history review

turned into a reasonably lengthy spell of employment, as in an account of 1305Ð06 from Kings Langley, Hertfordshire, where a garcio drove away rooks for 33 weeks at the winter and spring seedings and received one quarter for every 16 weeks for doing it.69 Otherwise, bird-scaring appears to have been occasional and probably dependent upon decisions made on the spot by officials, as at Bosham, Sussex, in 1302Ð03, where a garcio was paid a bushel of barley (for an unspecified time) to keep crows from a crop of beans Ôby order of the bailiffÕ.70

Another activity seemingly directed at young people as much as bird-scaring was harrowing, shown as the smallest grouping in Figure 2. Leading harrowing horses was often connected to bird-scaring, as at Aldenham, Hertfordshire, in 1298Ð99, where the harrower (herciator) also acted as a ÔrookherdÕ (that is, chasing away rooks) and received an equivalent of one quarter per 16 weeks for this dual role.71 Altogether, it was probably no accident that the crow-scarer and the harrower are shown together in Figure 4, taken from the Luttrell Psalter, and which was possibly meant to represent a child labour scene. The distribution of harrowers (herciatores) when rates of grain payments could be determined did, however, show a reasonably significant proportion getting liveries at the Ôfirst-tierÕ level: 21 of the total 61 harrowers, or 34.4 per cent, worked less than 16 weeks per quarter of grain (Appendix C, Part 2), so it is probable that they ranked higher than bird-scarers. Harrowing was sometimes connected with carting,72 and it seems that the natural progression of harrowers was probably to go onto helping and possibly eventually becoming carters, as in the case of the garcio Ôgoing to harrow and cartÕ at Hamstead (Marshall), Berkshire, from 1 November to 28 December 1298, again at the Ôsecond-tierÕ rate of one quarter per 16 weeksÕ work.73

The archetypal task for young people, though, was some form of herding or looking after animals generally. This could start with the herding of domestic geese, as in the case of the ÔgirlÕ (puella) who kept around 40 demesne geese for the payment of four and a quarter bushels of grain at Thorncroft, Surrey, for an unspecified time in 1310Ð11.74 Goose-herding might have involved people even younger than bird-scarers: of the 11 cases where a livery was given to a gooseherd, in only two was a rate indicated Ð one at one quarter per 16 weeks and the other at a quarter per 24 weeks.75

Care of sheep and particularly helping shepherds at lambing time also undoubtedly involved young people. A seemingly very young person, considering the low rate of pay, helped a

69 In liberatione unius garcionis fugantis frugellas de blado seminato ad utrumque seminem per xxxiij septi- manas, ij quarteria j pecka [here one-half bushel?], qui cepit quarterium ad xvj septimanas: TNA, SC 6/866/16, m. 1d.

70 In liberatione unius garcionis custodientis fabas pro cornicibus, j busellus, precepto ballivi: TNA, SC 6/1020/24, m. 2d. There were 27½ acres of beans sown this particular year.

71 In liberatione j herciatoris & eiusdem Rocherde per xviij septimanas tempore utriusque seminis, j quarte- rium j bussellus: WAM 26046.

72 As in the case of the Ôsub-carterÕ who also har- rowed at Berkhamsted, Herts., in 1296Ð7: L. Margaret

Midgley (ed.), Ministers’ accounts of the earldom of Cornwall, 1296–97 (Camden Soc. Third Ser. 66 and 68, 1942Ð5), I, pp. 20, 24.

73 TNA, SC 6/748/27, m. 1d.

74 MCL, 5742, m. 1d.

75 That is, at Brent, Somerset, in 1302Ð3, where the gooseherd was given a bushel for two weeksÕ work (Glastonbury Abbey Documents at Longleat; hence- foreward GAD; these are available on microfilm, which was used for this study Ð 11271, m. 3r). Also, at Ketton, Durham, in 1299Ð1300, the gooseherd was given two bushels for six weeksÕ work (DUL, DCD Enrolled Manors, 1299Ð1300, m. 3r).

shepherd for six weeks during lambing season for a ÔgiftÕ of two bushels, one of wheat and one of peas, equivalent to one quarter per 24 weeks (at Therfield, Hertfordshire, in 1307).76 A similar reference is found in a 1305Ð06 account for Pitchford, Shropshire, where a garcio Ôstood with the shepherd in lambing time and afterwards kept the lambs over a total of eighteen weeksÕ and was given the modest grain payment equivalent of one quarter per 24 weeks Ôfor [his] livery and wageÕ.77 Shepherding as a whole probably involved a wide range of adults and youngsters. Taking all sheep carers together (and not to be confused with the ÔshepherdsÕ in Figure 2, which only included those styled as bercarii), the total number of shepherding people, including ÔkeepersÕ, garciones, and occasional pages involved in sheep management presented a wide range of grain payments (see Appendix C, Part 2). Altogether, 84 of the total 464 sheep carers, or 18.1 per cent, were paid at a rate of one quarter per 16 weeks or less, indicating that nearly one in five of the people caring for demesne sheep was Ôsecond-tierÕ and, in this case, probably young.

Pigs, despite their reputation as difficult animals to control, as George Arnold (mentioned above) testified about his childhood in the nineteenth century,78 nonetheless seem even more likely to have been looked after by young minders. It is probably more than just literary fancy that the well-known late medieval outlaw tale, ÔAdam Bell, Clym of the Clough, and William CloudesleyÕ, twice characterized the town swineherd acting as a go-between for William Cloudesley and his wife as a Ôlytle boy(e)Õ.79 In the account sample, the distribution of grain payments for swineherds and keepers of pigs (Appendix C, Part 2) does have a broader spread across the first- to second-tier divide than, say, more obviously young people like harrowers and bird-scarers, but nonetheless the majority of them Ð 76 (or 60.8 per cent) of the 125 total Ð were in the Ôsecond-tierÕ group of having to work 16 weeks or more per quarter of grain.80 Swineherds would command a higher wage when the pigs were numerous, as at Elton, Huntingdonshire, in 1305Ð06, where a porcarius was given one quarter per 12 weeks for looking after a herd of pigs that totalled around 100.81 There were, however, many very poorly paid swineherds, often getting a quarter or less for the entire year.82 This raises the possibility

76 In dono cuidam garconi auxilianti bercarium tempore agnelationis per vj septimanas, j busellus fru- menti & j bussellus pisae: TNA, SC 6/872/17, m. 4r. The word dono instead of the more usual liberatione was perhaps meant to emphasize the one-time nature of the payment.

77 Et in liberatione j garcionis qui stetit cum bercario in tempore agnelationis & in posterum ad agnos cus- todiendes in toto per xviij septimanas, vj busselli pro liberatione & mercede: Lichfield RO, Cox reference no. G8, m. 1d.

78 Where he noted that looking after pigs Ômade him reflect with fondness on his earlier sheepÕ: Humphries, Childhood and child labour, p. 219; see also B. Gregory Bailey, Meaghan E. Bernard, Gregory Carrier, Cherise

L. Elliott, John Langdon, Natalie Leishman, Michal Mlynarz, Oksana Mykhed, and Lindsay C. Sidders, ÔComing of age and the family in medieval EnglandÕ, J. Family History 33 (2008), pp. 41Ð60 (esp. p. 54),

concerning the difficulties with pigs.

79 Stephen Knight and Thomas Oldgren (eds), Robin Hood and other outlaw tales (sec. edn., 2000), p. 246, lines 169, 173.

80 As in Appendix C, pt 2. The great majority of cases (119 of the 125) were styled simply as porcarii (swine- herds); in addition, there were three custodes porcarii and three garciones custodes porcarii. Three people combining swine herding with other duties are not included among the 125.

81 TNA, SC 6/874/12, mm. 1dÐ2d. These more senior swineherds are probably envisioned in the SeneschaucyÕs Ôoffice of the swineherdÕ: Seneschaucy, cc. 58Ð60, in Oschinsky, Walter of Henley, p. 285.

82 For example, the swineherds on the Glastonbury Abbey manors of Badbury and Idmiston, Wilts., and Greinton, Somerset, all in 1302Ð03, were paid two bushels, five bushels, and one quarter per year respec- tively: GAD, 11271, mm. 24r, 31d; GAD, 11246, m. 12r.

206

agricultural history review

that some of these low payments are because the swineherd was actually combining care of the demesneÕs pigs with care of pigs from others in the community, for which he was presumably also paid in some fashion, as implied in the case of an account for Bourton-on-the-Hill, Gloucestershire, for 1298Ð9, where the common swineherd for Bourton kept the pigs of the lord (the abbot of Westminster) with the other pigs of the community for a livery of six bushels for the year.83 Our supposition, though, is that, even with this complication, the weight of evidence of low wages Ð plus allusions by ÔAdam BellÕ and others, and later by George Arnold Ð lies with swineherds being mostly children.

The guarding of cattle and horses similarly seems to carry the same child or juvenile element to it, although the cases are not as frequent as for the minders of sheep or pigs. They were often connected to the summer or harvest period when working animals in particular would be idle, as in a 1298Ð9 account for Chilbolton, Hampshire, where a garcio was recorded as guarding the ÔaffersÕ (working horses), oxen, and (other) idle animals while Ôthe famuli harvested and did other worksÕ, at a livery that would have him working 16 weeks for a quarter of grain.84 Certainly, on the basis of pay, some of these summer livestock carers seem to have very young, as in a 1299Ð1300 Monkton Deverill, Wiltshire, account, where it was recorded that a garcio was given two bushels of grain for his Ôstipend and foodÕ for keeping the averia (a more general term for livestock as a whole) Ôfor the whole summerÕ; if the time covered was, say, eight weeks it would only give a rate of one quarter per 32 weeks.85

One thing that is clear about second-tier workers is the great variety of tasks they could be asked to do, as against first-tier staff who specialized in being ploughmen, carters, shepherds, and so on. Thus, a garcio given a quarter per 16 weeks was charged with Ômaking the pottage of the manorial servants, going to the harrowing, and keeping the birds away from the curia and the corn for the yearÕ at Cams, Hampshire, in 1301Ð02, for which he received a livery of one quarter per 16 weeks.86 Similarly a garcio protected the fields from birds, kept lambs after separation from their mothers and old sheep in summer at Easton in the same county in 1298Ð9, also receiving one quarter of grain per 16 weeks,87 while at Broadwell, Oxfordshire, a garcio made pottage for the famuli and Ôkept the fire and the courtÕ from 20 January to 12 May in 1308, a period of 16 weeks for the payment of a quarter, half of wheat and half of barley.88 All of these instances are reminiscent of the variety of tasks performed by William Arnold in the nineteenth century before he was seven. On the female side, an ancilla at Walton, Northamptonshire, in 1300Ð01, kept the manor, made the pottage for the famuli, winnowed corn, and loaded carts in the harvest, for which, over the year, she received a livery at a rate requiring a little over 20 weeksÕ work per quarter.89

83 In liberatione j porcarii de ville custodientis porcos domini cum aliis porcis de villa per annum, vj busselli: WAM 8249, m. 1d.

84 Hampshire RO, Dean and Chapter [hereafter HRO Dean & Chapter] of Winchester Muniments, Account Roll III, Chilbolton, 1298Ð99, m. 1d.

85 GAD 9685, m. 2r.

86 Page (ed.), Pipe roll … of 1301–2, pp. 364, 366.

87 HRO, Dean & Chapter, Roll III, Easton, 1298Ð99.

88 TNA, SC 6/957/6.

89 NRO, F(M) charter/2388, m. 6d.

figure 5: Luttrell Psalter spreading furrows scene

(© The British Library Board Add. MS 42130, fo. 171v)

(d) Second-tier workers: the elderly and others

‘Elderly’ is applied here to those people of advanced years who were perceived as no longer working as effectively as they did in their prime, to the extent that they were paid liveries at the second-tier rather than the first-tier rate. The age at which this would happen would clearly be variable, but in the medieval context certainly anyone 60 and above would qualify and, depending upon the person, their age might be considerably lower.90 We have already considered the possibility that the elderly had a presence among supervisory personnel, judging from occasional low rates of payment or the occasional use of subordinate terminology for these overseers (for example, the garcio at East Knoyle above). There were other tasks among second-tier workers that might be considered more typical of older people than, say, the young. One was gardening. Normally gardening would be done by first-tier staff.91 However, on some occasions, the task was more poorly paid, so that seven of the 30 people in the sample (23.3 per cent) who had gardening as all or part of their duties did so at the rate of one quarter per 16 weeks (Appendix C, Part 2). Again, perhaps these were elderly people who were entrusted with work of some responsibility but were given less because of their age. At Chilvers Coton in Warwickshire, in 1309–10, this seemingly elderly labour was juxtaposed with what was probably young labour, where a ‘cook/gardener’, who also tied sheaves, was preceded immediately in the record by a swineherd/harrower; both received a livery equivalent to one quarter (of mixed grains) per 16 weeks’ work.92

Another indication of the elderly among second-tier famuli initially comes not so much from the database, but from yet another illustration from the Luttrell Psalter (Figure 5), which shows a man and woman using long-handled mallets to break up large lumps of earth that

90 Men and women 60 years and above were exempted from the mid-fourteenth-century labour leg- islation in England, and the 20-year span from 40 to 60 was a period when workers were felt to experience a gradual decline in their physical capabilities: Shulamith Shahar, Growing old in the Middle Ages: ‘Winter clothes us in shadow and pain’, trans. from the Hebrew by Yael Lotan (1997), ch. 1 (esp. pp. 26–7); see also Youngs, Life cycle, pp. 163–5, for the age of 60 being considered decidedly ancient in the later Middle Ages.

91 As at Little Hinton, Wilts., where a gardener was given a substantial livery rate of one quarter per eight weeks for preparing a garden over 15 weeks: HRO, Dean & Chapter, Account Roll III, Little Hinton, 1298–9, m. 1d.

92 In liberatione j porcarii herciatoris & alia neces- saria facientis, iij quarteria ij busselli per annum. In liberatione j coci gardinarii & tassatoris per annum iij quarteria ij busselli: TNA, SC 6/1040/18, m. 1d.

208

agricultural history review

were still left after ploughing and harrowing. As Janet Backhouse has commented,93 the couple appears very elderly, with the man (when one looks closely) having a grizzled beard along his jaw line. The activity seems to be what the accounts call Ôspreading furrowsÕ (spargens sulcos), as suggested by an account for Feering with Pattiswick, Essex, in 1299Ð1300; here a livery was given to a garcio Ôspreading furrows and making water-furrowsÕ, the latter action indicating the clearing out of water channels and implying that Ôspreading furrowsÕ took place on the top of the ridge, as also suggested in Figure 5.94 Altogether in ourfamuli sample, there were nineteen individuals, found on 17 manors, who were given grain liveries for Ôspreading furrowsÕ as all or part of their activities. Of the 13 cases where a livery rate could be determined for these Ôfurrow-spreadersÕ, 12 were required to work 16 weeks for a quarter of grain, putting them firmly in the Ôsecond-tierÕ ranks. The thirteenth case was even more telling by involving the combination of Ôfurrow-spreaderÕ and bird-scarer at Lawling (near Mundon), Essex, in 1304Ð05. This individual had to work 32 weeks for his quarter of grain.95 This last case suggests that the furrow-spreader cum bird-scarer was perhaps young, while the more physically demanding work suggested at Feering with Pattiswick might suggest an older, and perhaps elderly, person. Interestingly both were called a garcio, and altogether seven of the 19 Ôfurrow spreadersÕ were styled as such. Also, these Ôfurrow spreadersÕ were concentrated very narrowly in one part of the country. Like bird-scarers, they had their greatest concentration in Essex, where 12 of the 17 communities with Ôfurrow spreadersÕ were in the county, with the other five being in the neighbouring counties of Suffolk (three cases) and Hertfordshire (two cases).96 The heavier soil conditions in this part of England might explain some of this, with a greater incidence of clumps of earth being left behind after ploughing and harrowing, especially in cooler, wetter conditions.97 Six of the 17 accounts mention the activity as taking place Ôin winterÕ and another two as being in the Ôwheat seedingÕ, the latter indicating the months leading up to Christmas. However, the tendency for the activity to be connected to bird-scaring might also suggest a greater concern to maximize grain production in an area close to the biggest urban centre in the country. It is notable, for instance, that other areas of England with heavy soils, notably the Midlands, did not record Ôfurrow-spreadersÕ at all.98

Other categories of second-tier workers are scarce in the sample. There was no one in the second-tier ranks of our sample with an obvious work-limiting disability, and indeed the only reference to someone with a disability of sorts concerned a Thomas Ôle harelippedeÕ (indicating at least a cleft lip), who kept the lordÕs wood at Ôla BereÕ for a part-year on the Winchester

93 Backhouse, Medieval rural life, p. 19.

94 In liberationej garcionis spargentis sulcos & facien- tis sulcos aquaticos apud Pateswyk & Haringgeslond per

ix septimanas, iiij busselli & dimidium (WAM 25601, m. 2d).

95 Cuidam garcioni spargenti sulcos & fuganti aves per viij septimanas ad seysonem yemale, ij [written over iiij crossed out] busselli (CCA, DCc Lawling 4, m. 1d).

96 Altogether two-thirds ofthe accounts for Essex (12 out of 18) had at least one Ôfurrow spreaderÕ. The manors of Feering with Pattiswick (Essex) and Clare (Suffolk) had two each.

97 For the generally heavy soil conditions in Essex, see H. C. Darby, The Domesday geography of eastern England (1952), fig. 55 (p. 217).

98 For the complicated interaction of soil type and commercial opportunities in determining agricultural practices, see John Langdon, Horses, oxen and techno- logical innovation: the use of draught animals in English farming from 1066 to 1500 (1986), pp. 255Ð6, 261Ð2; Bruce

M. S. Campbell, ÔTowards an agricultural geography of medieval EnglandÕ, AgHR 36 (1988), pp. 87Ð98 (esp.

p. 95).

Cathedral Priory manor of Barton Priors, Hampshire, in 1298–99. This required that he worked for just over ten weeks for each quarter of grain he received, a ‘first-tier’ rate signifying a capable, adult worker.99 More obviously handicapped by economic circumstances in the sample were the three and two paupers respectively on the Northamptonshire Peterborough Abbey manors of Kettering (in 1299–1300) and Cottingham (in 1309–10) who were recorded, in the first instance, as each receiving liveries of one quarter per 16 weeks over the entire year and, in the second, as getting an equivalent of about one quarter per 47 weeks (again over the entire year), while a single ‘pauper woman’ at Bewley, Durham, in 1299–1300 received a quarter of wheat over 16 weeks.100 In none of these cases was it indicated what exactly, if anything, these paupers did, so they may simply have been instances of charity.101

(e) Second-tier workers: numbers and proportions

We have presented evidence for the various types of second-tier workers found in our sample. Can we be more specific about their actual numbers and proportions? We are starting with the young first, since rigorous estimates of the extent of child and adolescent labour are hard to establish for this period, and some indication of its scale in this study would provide a useful starting point for discussions on the matter. Thus, in breaking down the 548 figure for the ‘second-tier’ element among the 3748 workers for which we have grain payment rates, if we subtract the apparent or probable women – the ancillae, mulieres, as well as ‘dairymaids’ making one quarter per 16 weeks or less (a total of 91, or 2.4 per cent of the 3748 total) – we are left with 457. If we further subtract supervisory personnel, gardeners, and ‘furrow spreaders’ receiving one quarter per 16 weeks or less – a total of 33 in our 3748 sample (or 0.9 per cent) – as being elderly (even though some of the furrow spreaders, in particular, might have been young), plus another ten to account for the poor (there were, as mentioned above, six in the sample) and possibly disabled, this would reduce the number of probable young to 414, or 11.0 per cent, of the 3748 total. On one hand, this might be considered a maximum, since there may have been older, lower status people involved (Harold Fox’s ‘landless males’, for instance102), but, on the other, given the probable larger representation of ‘second-tier’ personnel, many undoubtedly young, among the 833 people in the sample for whom a grain payment rate could not be ascertained, this percentage might well be on the low side and, in any case, a proportion of young of this size was almost necessary simply to replace some at least of the first-tierfamuli and to cover those jobs, like bird-scaring, that were probably only done by the neophyte in any case. Our conclusion at this point, then, is that the proportion of young people – they were probably predominantly male and we might put the top age at, say, 14, since 15 was the age that medieval males were considered adult enough to farm land on

99 In liberatione Thomae le harelippede custodientis boscum domini apud la Bere per xxiiij septimanas, ij quarteria iij busselli: HRO, Dean & Chapter, Account Roll III, Barton Priors, 1298–9, m. 1d.

100 In liberatione iij pauperum per annum, ix quarteria vj estrichae quasi capiunt quarterium per xvj ebdoma- das (NRO, F(M) Charter/2388, m. 15d); In liberatione ij pauperum per annum, ij quarteria j estricha dimidium

(ibid., Charter/2389, m. 27d); In liberatione cuiusdem mulieris pauperis per xvj ebdomodas, j quarterium fru- menti (DUL, DCD Enrolled Manors, 1299–1303, m. 1r).

101 For the term pauper as indicating a member of the ‘professional’ poor in medieval society, see Miri Rubin, Charity and community in medieval Cambridge (1987), p. 267.

102 Fox, ‘Exploitation’, passim.

210

agricultural history review

their own103 – was most likely in the 10–15 per cent range among the demesne famuli and that it tended to vary by region according to such things as population density, managerial policy, or urban demands upon agriculture. As detailed above, the types of employment entrusted to these ‘young’, from bird-scaring, harrowing, keeping guard over animals of all kinds, to being introduced eventually to the plough and cart, is entirely consistent with those tasks that seem to have been entrusted to younger people in agricultural societies generally, even those much closer to our own era.104

A maximum for the women in the sample can be obtained by assuming the 192 ‘dairymaids’ at both the first- and second-tier levels were all women (although some clearly were not) plus adding the 46 specifically designated women (ancillae, mulieres, and so on, as in Appendix C, Part 2), giving 238 individuals, or 6.4 per cent of the total 3748 famuli with known grain livery rates. Given the 10–15 per cent estimate for (mostly) young males above, it appears that these young males outnumbered women of any age and of any position among the famuli by around two to one. This 6.4 per cent figure is smaller than recently published figures for females in agriculture based upon the 1381 poll tax, which are around 50 per cent higher than the proportion of women found in this famuli sample,105 perhaps underscoring the longer- term, male-oriented nature offamuli employment overall. Young males dominated the elderly by even more, since those 33 cases stated above of males doing adult jobs but seemingly paid at a second-tier rate comprised 0.9 per cent in the 3748 total. Even if we add another ten people for the poor and possibly disabled, this makes 43, or 1.1 per cent, so that child and adolescent males would outnumber the elderly, poor, and possibly disabled by at least ten to one, but it does indicate that the elderly did have at least an occasional presence on demesnes and perhaps were seen as a steadying influence upon young male employees. The numbers of adolescent or child males, the elderly, and women taken together, however, were themselves dwarfed by the number of adult males who seemingly comprised at least 80–90 per cent of thisfamuli sample, and it must be emphasized again that these adult males worked for significantly longer periods and at higher livery rates throughout the accounting year.

III

This study has mapped out an English labour force from over 700 years ago. With its 105,000 or so workers (Appendix A), demesne famuli represented a reasonably sized proportion of English agricultural labour of the time, probably around ten per cent.106 Among other things,

103 As in the age of majority (15) for socage tenure: Nicholas Orme, Medieval children (2003), p. 327. In comparison, 90 per cent of boys were in work by age 14 during the Industrial Revolution: Humphries, Child- hood and child labour, fig. 7.1 (p. 177).

104 At the end of the twentieth century, 70 per cent of child labour was engaged in agriculture and related activities world-wide: Kaushik Basu, ‘The economics of child labour’, Scientific American 269, no. 4 (Oct. 2003), pp. 84–91 (esp. p. 87).

105 Stephen Broadberry, Bruce M. S. Campbell, and

Bas van Leeuwen provide a sample from the 1381 poll tax of 16,877 males and 1755 females working in agri- culture, a proportion of females of 9.4 per cent: ‘When did Britain industrialise? The sectoral distribution of the labour force and labour productivity in Britain, 1381–1851’, Explorations in Econ. Hist. 50 (2013), Table 1 (p. 17).

106 If the demesne proportion of all agricultural land across England was 20–25 per cent (Appendix A) and the famuli supplied a third to a half of the labour for demesne production (n. 5 above), then, if labour

what this study emphasizes is the considerable male-oriented nature of the famuli, above 90 per cent, certainly when compared to large farms of the early modern period where the proportion of women among servants seems to have averaged around 25 per cent.107 Furthermore, as dairying in particular became more gender-uncertain from the beginning of the fourteenth century, it appears that males were set to dominate even more powerfully among the famuli during the rest of the century.

Perhaps the most novel contribution of this article is to attempt a more exact breakdown of this workforce by stage in life without any direct information as to the age of workers. Consequently, the 10Ð15 per cent of child and adolescent labour among the famuli approx- imated here through inference from grain livery payments and job descriptions is at least conceivable in the circumstances, even if it is, say, less than the proportion of child labour in many parts of the world today.108 Elderly labour is more difficult to tease out, but the one per cent or so calculated above (and including the poor and possibly disabled) must easily stand as a plausible minimum. At the very least, these estimates provide a point of reference from which to compare age-related labour analyses from other sources, periods, or countries.

Finally, it is important to re-emphasize that this ÔsnapshotÕ presents a pre-eminently static view. The makeup of thefamuli, however, was anything but static. The appearance of the labour of women and the young in particular seemingly fluctuated according to the health of the overall economy.109 The methodology here of presenting the evidence in a bipartite first- and second-tier fashion, regardless of the degree of confidence readers might have about this division and the absolute figures generated, does allow useful comparison over time. The surviving manorial account material from which this examination of the famuli c.1300 was drawn is exceedingly plentiful for at least the period c.1270 to c.1420,110 a range conveniently having the initial advent of the Black Death at or near its central point. In this regard, one thing that might strike readers is that the level of Ôsecond-tierÕ personnel posited here for the beginning of the fourteenth century seems high, especially for the young, in a time of supposed labour glut. We feel, indeed, from other evidence not presented here that employment was reasonably robust at the time and only began to falter in the decades immediately preceding the advent of the plague.111 An

Note 106 continued

productivity was roughly equal across all sectors and workers (perhaps a debatable Ôif Õ: see n. 6), then the proportion that the famuli represented in total agricul- tural labour would range from 6.7 [1/5 × 1/3 × 100] to 12.5 [¼ × ½ × 100] per cent.

107 Jane Whittle, ÔHousewives and servants in rural England, 1440Ð1650: evidence of womenÕs work from probate recordsÕ, Trans Royal Hist. Soc., sixth ser., 15

(2005), pp. 51Ð74 (esp. p. 57). We are grateful to one of the anonymous referees for referring us to this very useful article.

108 As in current sub-Saharan Africa, where an esti- mated 29 per cent of children from 5 to 14 years of age work for a living; the figure for Asia is 19%: Basu, ÔEconomics of child labourÕ, p. 90.

109 As their presence (or not) in royal works accounts

indicates: Langdon, ÔMinimum wagesÕ.

110 Slavin, ÔSourcesÕ, p. 133.

111 A preliminary examination of accounts for Eybury, Middx (the home farm of the Abbey of West- minster) from 1275Ð6 to 1346Ð7 shows that, although womenÕs employment held up fairly well among the famuli over this period, young male labour seemingly shrank considerably, to the point that the total grains dispersed to these younger workers in 1346Ð7 were only a quarter of what they collectively received in the late thirteenth century: WAM 26853Ð26902. For similar results from royal works accounts, see Langdon, ÔMinimum wagesÕ. Possible theoretical reasons for this, pitting the metrics of individual real wages against family income, may be found in Langdon and Mass- chaele, ÔCommercial activityÕ.

212

agricultural history review

examination of the famuli over the longer sweep from the thirteenth to the fifteenth centuries will help us examine further this apparent incongruity while revealing considerably more about the nature of labour during a time of extraordinary economic and social transformation.

Appendix A:

The numbers of Famuli across England, c.1300

The size of the famuli workforce across England around 1300 can only be an estimate, but we do have some data from which to judge the matter. There are three things to be considered here: a) the total yearly sown acreage that demesnes encompassed; b) the portion of that sown land probably worked by famuli, from which their number can be determined; and, further, c) the proportions of service versus stipendiary famuli.

a) The most recent authoritative assessment of the total annual sown land in medieval England in 1300 puts it at 8.16 million acres.112 Campbell, in his country-wide update of Kosminsky’s figure that 32 per cent of land was in demesne (based upon the 1279 Hundred Rolls for a number of Midland and eastern counties), downgraded the demesne portion to a quarter or even a fifth.113 Using the more conservative of these figures – a fifth – then the amount of land sown each year on demesnes would be 8.16 million acres × 0.2 = 1.63 million acres.

b) How much of this land would be serviced by famuli labour as against customary labour services or ‘on the spot’ hiring of workers for specific tasks? Here we can use the employment offamuli ploughmen, both holders and drivers as shown in Figure 3, as an indicator. Many accounts in the sample provided very clear information both about the full set of famuli ploughmen and the sown acres for that account year. Using the information from 116 such accounts (78 ecclesiastical and 38 lay),114 the mean number of sown acres per full-time famuli ploughman was 46.8 (median, 45.2).

It is traditional to assume that each plough could handle 120 acres per year.115 Keeping this in mind, and assuming two ploughmen per plough,116 the mean sown acres per demesne plough cultivated from the 116 accounts above would be 46.8 × 2 = 93.6. This would leave, on average, 120 – 93.6 = 26.4 acres of unused capacity for each plough, which might have been used, say, for ploughing up fallow, although the ploughing facilities on some estates must have been hard-pressed to cover even the sown acreage.117 It seems most probable that, as a

112 Stephen Broadberry, Bruce M. S. Campbell, Alex- ander Klein, Mark Overton and Bas van Leeuwen, British economic growth, 1270–1870 (2015), Table 3.03 (p. 89) in the ‘Total sown’ column.

113 E. A. Kosminsky, Studies in the agrarian history of England in the thirteenth century, trans. R. Kisch, ed. R. H. Hilton (1956), pp. 89, 91; Campbell, English seigniorial agriculture, 1250–1450 (2000), p. 58.

114 These were drawn from the ecclesiastical estates

of Westminster Abbey, Canterbury Cathedral Priory, Norwich Cathedral Priory and Peterborough Abbey and the lay estates of Henry de Lacy, earl of Lincoln, and Roger Bigod IV, earl of Norfolk.

115 E.g., Campbell, English seigniorial agriculture, p. 121.

116 In this smaller 116 manor sample, there was only one case of a plough being handled by a single person, that is, for a small horse plough at Thornham, Norfolk, in 1299–1300 (Norfolk RO, DCN 60/37/9); the other

115 ploughs had at least some oxen drawing them and required both a holder and driver.

117 As on seven manors of Westminster Abbey, which had a mean of 129.8 sown acres per demesne plough. This would mean that, on average, at least 9.8 sown acres would probably have been performed by other ploughing sources.

general policy, demesne ploughs were directed towards the more critical ploughing of ground to be sown, while the ploughing of fallow was left for customary ploughing services or hired ploughing, as David Stone has suggested for the manors of the bishop of Ely.118

Consequently, based on the above, since famuli ploughmen seem to have been numerous enough on average across estates to plough the demesne sown acreage at least, then a minimum of their total full-time numbers might be determined by simply dividing the countrywide figure of 1.63 million demesne acres to be sown each year by the mean sown acreage per ploughman given above (46.8) – that is, 1,630,000/46.8 = 34,829, rounding off to, say, 35,000 full-time famuli ploughmen. Converting this to a figure for the famuli as a whole, the 1423 ploughmen in our sample worked for a total of 65,520.4 weeks compared to 157,974.4 for the 3748 famuli as a whole. If we scale up from the 35,000 ploughmen figure above, this would give (35,000 × 157,974.4)/65,520.4 = 84,388. These, however, are full-time equivalents. Since the total 3748 workers only averaged 42.1 weeks per year, the actual number of people working, full-time and part-time, would be (84,388 × 52)/42.1 = 104,232. Since these calculations did not include the 833 people in the sample for whom we could not determine grain payment rates, who probably worked fewer weeks in the year than the average and thus would reduce the 42.1 denominator in the previous calculation, a minimum of 105,000 to include these people would again seem plausible. The contextual relevance of this number is perhaps best appreciated when comparing it to the estimated 600,000 smallholding families across England.119 Famuli employment, if confined to this class (a big ‘if’, since it is hard to know how to account for the landless, including those ‘drifting down’ from better-off peasant classes but still contributing earnings to their families), would be a significant but certainly not overwhelming contribution to the overall well-being of smallholders.120 In that sense, the first priority is probably best to think of the famuli as forming independent households in their own right, as we attempt to do in Appendix B.

c) Concerning the service famuli proportion among this 105,000 total, this is estimated from the service ploughmen recorded in the ‘acquittances’ sections of the accounts for the estates of the bishop of Winchester, Winchester Cathedral Priory, and Glastonbury Abbey, which provided most of the service famuli recorded among the documents examined in our study (see note 13 above). The number of such ploughmen came to 436. If we round this up to 500 to include all the possible service famuli in our account sample and add the 4581 stipendiary famuli that have been the prime focus of this study, this would come to a total of 5081, of which the service element would be 9.8 per cent and the stipendiary 90.2 per cent. Applying these percentages to the estimated total of 105,000 famuli above would result in 10,290 servicefamuli and 94,710 stipendiary ones in England c.1300.

118 Stone, Decision-making, p. 70.

119 Christopher Dyer, Standards of living in the later Middle Ages: social change in England, c.1200–1520 (revised 1998 edn), pp. 126–7.

120 Having said this, it does appear, on admittedly

slim evidence, that smallholders were a very important source for recruiting famuli: P. D. A. Harvey, A medi- eval Oxfordshire village: Cuxham, 1250 to 1400 (1965), pp. 77–8; see also n. 146 below.

214

agricultural history review

Appendix B:

Total remuneration offamuli

What we attempt here is to give a sense of total remuneration in terms of kilocalories (hereafter ‘kcal’ or ‘kcals’) for typical first- and second-tier famuli working for a full year of 52 weeks. We are going to consider stipendiary workers only, so that reductions or ‘acquittances’ of rent given to service famuli are not involved here.

(a) Grain Liveries

In terms of assessing the generosity or not of the grain liveries for a first-tier worker, we will use the median livery rate for first-tierfamuli of one quarter (= eight bushels) per 12 weeks’ work (Figure 1), which gives an annual grain payment of 4.33 quarters, or 34.6 bushels. To assess kcal equivalents, we propose a range, based on, first, rye, as the upper bound, and second, a mixture of barley and oats, as the lower one. The caloric equivalents of a bushel of these two options would be 77,520 and 63,564 kcals respectively,121 so that 34.6 bushels would yield a range of 2.68 million kcals (rye) to 2.20 million kcals (barley and oats) . Campbell put the kcal extraction rate writ large for all grains c.1300 at 58 per cent (including the use to which it was put, from pottage through to brewing, as well as loss through vermin and rotting).122 If we apply this to the kcal equivalent range above, then the net result would be 1.55m (rye) to 1.28m (barley and oats) kcals. Campbell also put the daily grain kcal requirement at 1500 per person, balancing the differences between male and female, young and old, and the fact some protein from meat and/or dairy products would be added for a reasonably healthy diet.123 Thus, a rate of one quarter per 12 weeks’ work at 1500 kcals could support 2.8 (rye) to 2.3 (barley and oats) persons over a year.124 If a more generous grain kcal per person per day was felt to be necessary, say at 2000,125 then the range would be reduced to 2.1 to 1.7 persons. This indicates that the most common grain livery rate was, in terms of sustenance (and overlooking cash wages for the moment), only just able to support an adult couple if at all.

Not surprisingly the outlook was even gloomier for second-tier famuli. From Figure 1 the median figure for the second-tier personnel was at the one quarter per 16 weeks’ work level (3.25 quarters, or 26.0 bushels, per year). Working from the assumptions above, the grain liveries for this group would have supported from 2.1 to 1.8 persons at the 1500 kcals per person per day requirement and 1.6 to 1.3 people at 2000 kcals per person per day.

(b) Cash stipends

The money wages that famuli received could ameliorate this situation, of course. Limiting ourselves to those for whom cash payments per year were stated or could be calculated in the sample – 1638 for the first-tier group and 115 in the second-tier – the median annual cash payments for both groups were 4s. and 2s. 6d. respectively. If these payments were converted

121 Campbell, English seigniorial agriculture, Table 5.04 (p. 215).

122 Ibid., esp. pp. 397–9.

123 Ibid., pp. 401–2.

124 For example, for rye the calculation was 1,550,000/ (1,500 × 365) = 2.8311.

125 Along the lines of, say, Dyer, Standards of living, pp. 134–5.

into grain, using Farmer’s prices for the first decade of the fourteenth century,126 4s. if spent on, say, rye would raise the amount of grain for consumption by 7.7 bushels (or 22.2 per cent by volume over the 34.6 bushels that a famulus/famula would get at a rate of one quarter per 12 weeks’ work). If spent on the barley/oats alternative, it would raise the livery by 10.1 bushels (or 29.2 per cent), an improvement over rye that was also reflected in kcal terms (by 7.6 per cent: 641,996 versus 596,904 kcals). Indeed, in purely sustenance terms, the optimal strategy for a first-tier famulus/famula receiving rye for their livery would be to spend their 4s. cash stipend on something like a barley/oats mixture (as long as the grains were not used for less efficient purposes like making ale), which, at a total of 3.32m kcals (2.68 + 0.64) and following the calculations above (including an extraction rate of 0.58), would support a range from 3.5 (at 1500 kcals per day per person) to 2.6 persons (at 2000 kcals per day per person).

Applying the same calculations to a second-tier famulus/famula receiving one quarter of rye per 16 weeks’ work – thus setting an upper bound for this category of worker – 2s. 6d. at Farmer’s prices for the first decade of the fourteenth century would buy 6.3 bushels of a barley/ oats mixture or an extra 400,453 kcals, making 2.41m kcals in all (that is, added to 2.01m kcals from 3.25 quarters of rye), supporting a range from 2.5 (at 1500 per person per day) to 1.9 persons (at 2000 kcals per person per day).

In short, even the most optimistic conditions above only gave sustenance for an equivalent of 3.5 people, perhaps a couple and three children, assuming the latter combined amounted to 1.5 ‘persons’. Such a fixation on food would, moreover, leave nothing for clothing, shoes, housing, or utensils (or, even if they made some of these themselves, cloth, leather, wood, and metal). Indeed, if one views famuli wages from another perspective, converting all grain payments to cash, even a first-tier famulus/famula would receive barely 1d. per day, while second-tier workers would receive around ¾d. per day,127 very much endorsing Dyer’s pessimistic view of the famuli existence.128

(c) Perquisites

There were, however, perquisites offered by famuli employment that would help to soften these realities or at least provide insight as to how life at these remuneration levels could be sustained. One was the likelihood that the famuli received a portion of ‘pottage’, or porridge,

126 Farmer, ‘Prices and wages’, p. 733, where the price of rye over the decade is given as 4s. 2d. per qr, barley at 4s. per qr, and oats at 2s. 4d. per qr. A 50-50 barley-oats mixture would theoretically be 3s. 2d., which was used here.

127 Assuming rye, probably the best grain to be given to the famuli, the 4.33 quarters that a worker at one quarter per 12 weeks’ work when converted to cash (based upon Farmer’s prices for rye in the previous note) would be 4.33 × 50d. (4s. 2d.) = 216.5d. Adding to this the median 48d. (4s.) cash payment received by such a worker, this would come to an annual ‘wage’ of 264.5d. If we assume 260 days of work per year, around the average used by Dyer for his construction

of medieval English wage-earning budgets (Standards of living, p. 226), this would come to an equivalent of slightly more than a penny per day, which at the beginning of the fourteenth century was a remunera- tion more consistent with that for women and the young (e.g., Langdon, ‘Minimum wages’). For second- tier workers, even with the supposition that they would be receiving rye for the 3.25 quarters (= 26 bushels) per year they earned at a livery at one quarter per 16 weeks’ work, which would be worth 162.5d. at Farmer’s prices, plus a second-tier median annual cash payment of 2s. 6d. (30d.), this would only amount to a total of 192.5d., or ¾d. per day, at a 260-day work year.

128 Dyer, Standards of living, p. 133.

216

agricultural history review

made of oats and/or peas, possibly to start the day or as snacks to support their exertions thereafter. As the numerous references to second-tier men and women making pottage for the famuli in the main text suggests, the practice was reasonably common and perhaps even ubiquitous, even if it did not always make it into the record,129 with each famuli being allocated an equivalent of 1–1½ bushels of oats/peas over the year.130 Such pottage was probably seasoned with salt,131 and, in one instance, it was indicated that it was prepared in an earthenware pot or bowl held over a fire by a tripod.132 This pottage, if shared equally, could add around five per cent to the sustenance for a first-tier worker and perhaps something around seven per cent for a second-tier one.133

The second common benefit beyond grain liveries and cash stipends for the famuli was the provision of feasts to celebrate important holidays, for which expenses were paid by the lord, at about 1½d. per feast per famuli member, as well as often a tip or gratuity (oblatio) of a halfpenny or a penny per person per feast. These relations-improving exercises between lord and employees were particularly common on the estates of Westminster Abbey, where two-thirds of manors – usually the larger ones – had at least one such feast per year, particularly at Christmas, but often at Easter and occasionally at other times as well, such as St. Michael (29 September) and All Saints (1 November).134 It is difficult to say how important these feasts were in a nutritional sense, because, among other things, it is not certain how many meals they entailed. A Battersea, Surrey, account for 1299–1300 indicates that the Christmas expenses for the famuli stretched over three days, and the reasonably generous ‘expenses’ of 12s. (itself written over 14s. crossed out) for the 20 or more famuli, which, over three days, would yield around 2.0–2.5d. per day per person, gives credence to what might have been a lengthier spell of banqueting and carousing,135 but the more normal 1½d. per feast per person mentioned above

129 Only seven of the 57 manors of the bishopric of Winchester, for instance, gave very clear indication of it: Page (ed.), Pipe roll … of 1301–2, pp. 52, 141, 274, 280, 341 (oats pottage) and 75, 199 (peas pottage). A probable eighth case (of oats pottage) was recorded for Adderbury, Oxon., where an entry in the oats section noted, ‘In making meal, 1 qr’, which seems to have been bound for the famuli (ibid., 150) and shows how easily such pottage might elude detection in the records.

130 The eight cases above (including Adderbury) indi- cated that 73.5 bushels of oats/peas were given to 52 identifiable stipendiary famuli, for a portion of 1.41 bushels each.

131 As at Turweston, Bucks., in 1299–1300, where three bushels of salt was bought for 12d. for ‘the pottage of the famuli and the dairy’: WAM 7761, m. 1r.

132 As at Witney, Oxon., in 1301–2, where the pur- chase of a pot and tripod was recorded for making the famuli’s pottage for work they were doing ‘in the park’: Page (ed.), Pipe roll … of 1301–2, p. 137 (under ‘Small Expenses’).

133 Assuming the 1–1½ bushels were oats and a 90% extraction rate – since, even though this oats was given

as pottage, some would likely be lost through milling or wastage – this would give a range of 54,302 to 81,454 kcals (using the kcal per bushel figure for oats in Camp- bell, English seigniorial agriculture, p. 215). At the 1.55m to 1.28m kcal range for the liveries after overall extrac- tion at 58 per cent above, depending upon the grain (rye versus barley/oats mixture), for the livery of afamulus/ famula at the one quarter per 12 weeks’ work rate, then the extra kcal percentage of this pottage would range from a minimum of 3.5 [(54,302/1,550,000) x 100] to a maximum of 6.4 [(81,454/1,280,000) x 100] per cent. For those at one quarter per 16 weeks, receiving annual liveries comprising, after 58 per cent extraction, 1.17m kcals (rye) and 0.90m kcals (barley/oats), the improve- ment would range from 4.6 to 9.1%.

134 At Hampstead, Middx, in 1289–90, there appears to have been such feasts at Christmas, Easter, and Mich- aelmas (WAM 32405, m. 1r), while at Oakham, Rutland, in 1299–1300 the feasts were at Easter and All Saints (WAM 20228, m. 2r). These references are to be found in the ‘Small Expenses’ (Minutae Expensae) part of the accounts, as are the references to various feasts below.

135 WAM 27504, m. 2r.

suggests that usually only a single dayÕs feasting was involved. In terms of total sustenance over a year, these celebratory feasts probably represented only a few daysÕ nutrition, even if the famuli gorged themselves and took away food for future consumption. They might also, however, have been enhanced by food provided for harvest and other customary ÔboonsÕ that famuli/famulae attended.

Where the information about holiday feasts is sometimes particularly useful, however, is in revealing more fully the working groups that comprised the famuli, often involving otherwise unrecorded members. Thus, in our sample, in a 1298Ð99 account for the AbbeyÕs manor at Aldenham, Hertfordshire, those attending the Christmas and Easter feasts were recorded as Ôthe reeve, the beadle, one carter, four famuli ploughmen, their helper (garcio), one cowherd, his helper, one shepherd, his helper, one smith, his helper, [&] a dairymaid [and] her [female: ancilla] helperÕ.136 None of the ÔhelpersÕ in this passage seemingly appeared elsewhere in the account and suggests broader family involvement among these famuli that might impart economies of scale that would help ameliorate difficult economic conditions.137

Our sources never clearly indicate whether members of the famuli received housing benefits as part of their remuneration. Some seem to have lived nearby, often on a smallholding,138 although Harvey indicates a substantial proportion may have had lodgings within the curia, the manorial range of buildings.139 It might be, too, that famuli could have benefits from, in effect, leasing or loaning the livestock and equipment held in the curia,140 as Harvey has argued for Cuxham, Oxfordshire.141

In short, the value offamulus employment should not be judged solely on the grain and cash payments that they received. Even if, c.1300, individual employment as a famulus or famula might be short-term, as Richard Britnell has suggested for the later fourteenth century,142 famulus positions, particularly at the first-tier level, seem to have been very solid and attractive jobs that lordsÕ officials could easily fill when vacancies arose. The attractiveness of such

136 In expensis prepositi Bedelli j carectarii iiij famu- lorum carucarorum garcionis eorundem j vaccarii gar- cionis sui j bercarii garcionis sui j fabri garcionis sui daye ancillae suae diebus Nativi domini & Paschae iijs. iiijd. [this amount was written over iiijs. crossed out]. In oblationibus eorundem xd.: WAM 26046, m. 1d. For an equally detailed example for Launton, Oxon., in 1289Ð90, see Bailey et al., ÔComing of ageÕ, p. 51.

137 For an indication of how this might work see the case of a famulus miller running the double watermill at Feering, Essex, where intense family involvement could certainly have altered what appeared from the perspective of a single employee an insupportable, not to mention exploitative, situation: Langdon and Mass- chaele, ÔCommercial activityÕ, pp. 69Ð70; the case is also discussed in John Langdon, Mills in the medieval economy: England 1300–1540 (2004), pp. 238Ð40.

138 See n. 146 below; also Harvey, Medieval Oxford- shire village, pp. 77Ð8.

139 Ibid., p. 77. The care with which the buildings were

kept might also suggest that at least some of the famuli resided there: see the discussion of mulieres, ancillae, etc., doing housekeeping for the curia above.

140 Livestock holdings were extensive on demesnes, as any perusal of manorial accounts will reveal: e.g., Page (ed.), Pipe roll of … 1301–2, esp. pp. 20Ð1, 24, 28, 32, 37Ð8, 45, etc. Equipment is often revealed in ÔutensilsÕ or Ôdead stockÕ sections at the end of accounts: e.g., ibid., pp. 15, 46, 54, 57, 61, 71, etc.

141 Harvey, for instance, cites a 1356 case where the famuli were allowed to use the demesne ploughs to plough their own lands before they attended to the demesne itself: Medieval Oxfordshire village, p. 77. The flexibility of the cowherdÕs access to the milk of the animals points in the same direction (see the Husbandry excerpt above in the section discussing dairymaids).

142 Where employment at Houghall, Durham, about a century later was normally on six-month contracts: see

n. 16 above.

218

agricultural history review

positions would increase even more if second-tier jobs could be filled by other family members, as Dyer suggests.143 The intensity – or seasonality – of employment might a factor here. If ploughmen really did plough less than 120 sown acres a year, as Appendix A suggests, it might leave much time to attend to other personal business while still enjoying an annual ‘salary’. The seasonal interplay between dairymaids and cowherds, with the former seemingly more active in the summer and the latter in the winter (as mentioned in the section about dairymaids in the main text), might suggest the same, although this alternation of slack and busy periods was not something that would necessarily apply across the famuli as a whole – shepherds in particular were probably busy with their sheep all year round.

Nevertheless the possibility of creating little ‘family businesses’ from famulus positions was certainly an option, as perhaps most obvious in the case of cowherds who leased demesne herds. We are only at the beginning of working out the mechanics of such ‘enterprises’, but they can certainly alter our perception of periods that are often characterized as being increasingly wretched for the great majority of people.144 These accommodations are easiest to perceive with stipendiary famuli, where clues as to supplementary income both on an individual and family level can at least be discerned.145 Indeed, the hardest to explain are the ‘service’ famuli, whose rent reductions of only a few shillings seem very difficult to square with the amount of work they were expected to do on demesnes.146 If the customary right to hold their land was involved, then their demesne ploughing, say, would be little more than another form of labour service (albeit using demesne livestock and ploughs) with the efficiency issues that involved: see note 6 above. In any case, this conundrum will have to remain a topic of future research and consideration.

143 Dyer, Standards of living, p. 133.

144 E.g., see Hatcher and Bailey, Modelling the Middle Ages, pp. 43–8; a more optimistic, family-oriented view is in Langdon and Masschaele, ‘Commercial Activity’.

145 In a few cases parcels of land held by famuli are revealed: Harvey, Medieval Oxfordshire village, pp. 77–8; see also the following note.

146 It may be little wonder that such ‘service’ person- nel would prefer a shift to stipendiary status, as hap- pened to a ploughman at Milton Podmore, Somerset, in 1302–3, who had 2s. 6d. relief from the rent of a ‘ferdell’ (a quarter-virgate, likely of around 7–8 acres) during the 30 weeks from St Michael to Hockday, that is, 29 Sept. 1302 to 16 Apr. 1303, in which case the relief

was worth 1d. per week. From then to the following Michaelmas he was put ‘to a livery and stipend’, which yielded him 1.8 quarters of wheat and rye (at a rate of one quarter per 12 weeks’ work) and a stipend of 2s. 4d. over these remaining 22 weeks. Pricing an assumed 50-50 split of the wheat and rye using Farmer’s data for the early fourteenth century (Farmer, ‘Prices and wages’, p. 733) and adding the result to the 2s. 4d. cash payment, gives a rate of 6d. per week, in this case an apparent and very impressive six-fold advantage for the stipendiary over the service option: GAD 11246, m. 21r–21d. In this case it might be possible that the ‘ploughman’ surrendered his holding in toto for his stipendiary famulus position.

Appendix C

part 1 : Total and regional percentages offamuli in various livery rate groupings

Livery Rate All famuli North Thames Basin South and Midlands (Weeks/ South West quarter) % n % n % n % n % n										East Anglia
										% n
5-7 3.3 125 0.9 3 9.5 121 0.1 1 – –										– –
8-9 14.2 531 0.6 2 25.9 329 18.1 149 4.3 31										3.3 20
10-11 21.6 811 9.5 31 25.1 319 23.1 190 11.6 84										30.7 187
12-13 43.1 1616 58.8 191 19.7 250 42.6 350 69.2 500										53.4 325
14-15 3.1 117 10.2 33 3.0 38 1.6 13 2.1 15										3.0 18
16-17 10.8 406 12.6 41 14.2 180 11.3 93 6.6 48										7.2 44
18-19 0.5 20 0.3 1 0.6 8 0.7 6 0.6 4										0.2 1
20-21 0.5 19 0.3 1 0.4 5 0.4 3 1.0 7										0.5 3
22-23 0.2 9 0.6 2 0.1 1 0.4 3 0.3 2										0.2 1
24-31 1.7 62 4.6 15 0.8 10 0.7 6 3.3 24										1.1 7
32+ 0.9 32 1.5 5 0.6 8 1.0 8 1.1 8										0.5 3
Total 99.9 3748 99.9 325 99.9 1269 100.0 822 100.1 723										100.1 609
Source: Authors’ manorial accounts database for 1289-90 to 1310-11. Counties for each region as are as specified in the note for Table 1. part 2 : Percentages of various types offamuli in various livery rate groupings (in order as discussed in text)
Livery Rate	Ploughmen		Carters		Shepherds		Cowmen		Dairymaids
(Weeks/Quarter)					(Bercarii only)		(Vaccarii)		(Dayae/Daiae)
	%	n	%	n	%	n	%	n	%		n
5-7	5.3	76	4.0	17	1.2	4	3.0	4	0.5		1
8-9	15.5	220	18.7	80	12.7	44	11.3	15	3.6		7
10-11	23.5	334	25.1	107	28.6	99	18.0	24	19.3		37
12-13	52.4	746	47.3	202	48.0	166	41.4	55	40.6		78
14-15	2.2	31	1.4	6	4.3	15	9.0	12	9.4		18
16-17	0.8	12	2.6	11	4.9	17	15.0	20	19.8		38
18-19	0.1	1	0.5	2	–	–	–	–	0.5		1
20-21	–	–	–	–	–	–	–	–	1.6		3
22-23	0.1	1	–	–	–	–	–	–	1.0		2
24-31	0.1	2	0.2	1	0.3	1	2.3	3	1.6		3
32+	–	–	0.2	1	–	–	–	–	2.1		4
Total	100.1	1423	100.0	427	100.0	346	100.0	133	100.0		192

220

agricultural history review

part 2 : continued

Livery Rate	Women		Pages		Garciones		Harrowers		All Sheep Carers
(Weeks/Quarter)	(Ancillae, Mulieres etc)		(Pagetti)						(including Bercarii)
	%	n	%	n	%	n	%	n	%	n
5–7	–	–	–	–	–	–	4.9	3	0.9	4
8–9	–	–	–	–	1.3	2	3.3	2	10.8	50
10–11	–	–	4.8	1	2.0	3	4.9	3	26.7	124
12–13	10.9	5	–	–	5.9	9	16.4	10	40.1	186
14–15	2.2	1	–	–	5.2	8	4.9	3	3.4	16
16–17	43.5	20	42.9	9	60.8	93	57.4	35	15.3	71
18–19	–	–	–	–	5.2	8	6.6	4	0.2	1
20–21	6.5	3	–	–	2.0	3	1.6	1	0.2	1
22–23	2.2	1	–	–	1.3	2	–	–	0.2	1
24–31	10.9	5	47.6	10	13.1	20	–	–	2.2	10
32+	23.9	11	4.8	1	3.3	5	–	–	–	–
Total	100.1	46	100.1	21	100.1	153	100.0	61	100.0	464

Livery Rate	Swineherds and		Messores		Gardeners		‘Furrow Spreaders,
(Weeks/Quarter)	‘Keepers, of Pigs
	%	n	%	n	%	n	%	n
5–7	0.8	1	–	–	–	–	–	–
8–9	7.2	9	19.5	15	13.3	4	–	–
10–11	6.4	8	10.4	8	20.0	6	–	–
12–13	19.2	24	53.2	41	36.7	11	–	–
14–15	5.6	7	3.9	3	6.7	2	–	–
16–17	42.4	53	9.1	7	23.3	7	92.3	12
18–19	0.8	1	–	–	–	–	–	–
20–21	4.0	5	1.3	1	–	–	–	–
22–23	0.8	1	1.3	1	–	–	–	–
24–31	8.0	10	–	–	–	–	–	–
32+	4.8	6	1.3	1	–	–	7.7	1
Total	100.0	125	100.0	77	100.0	30	100.0	13

Source: Authors’ manorial accounts database for 1289–90 to 1310–11.