Introduction: Research Areas

1. Literacy Rates

A further indication of the status of those who swore is provided by the literacy levels of the oath-takers. It was expected that subscribers would sign their names, or else make a mark if they were illiterate. That there was some variety in the levels of literacy has already been discussed. The oath rolls also provide some scope for estimating literacy levels across the county. However, the literacy data is not consistent throughout the rolls and so they do not provide a suitable source for conducting detailed research into patterns of literacy in early eighteenth century Devon. In particular the practice of initialling the documents that became increasingly common from September onwards may distort the figures. As noted above, it is not universally apparent whether initials were subscribed by the individual taking the oath or the court clerk. In many cases the initials appear to have been formed by an individual with sound writing ability. This problem is particularly acute in the case of individual court sessions where as many as 90% of the signatories ‘marked’ the manuscript with an initial.¹¹⁵ On 23 November 181 of the 245 men and women who took the oath at the Swan in Silverton ‘signed’ an initial against their name. Figure 9 shows a sample of the marks made by oath-takers on this occasion. As can be seen, the letters are often neatly formed, suggesting that the person responsible for writing them was capable of writing much more than a single letter. However, it cannot be determine whether the person responsible for such ‘marks’ was the oath-taker themselves or the scribe or clerk of the court. In some cases, the ‘mark’ appears to be in the same hand as the scribal entries, for example William Turner, Henry Vaulter, and Susanna Taylor.

This evidence presents serious problems if the oath rolls are to be used as an indication of literacy in early Georgian Devon. It simply cannot be determined whether the practice of initialling beside individuals’ names was caused by their inability to write or by a desire on the part of all concerned to speed along what would have been a decidedly lengthy process. That said, even in cases where 90% of those present marked with an initial letter, such as at West Teignmouth in December, a small proportion did sign their names. Therefore, it is to be assumed that this option was available to all of those present and was declined either because they could not write their name in full or they preferred not to do so on this occasion. Considering that for some people writing would have been a difficult activity – perhaps they could only write their name and nothing else – it should not be assumed that the failure to sign the oath rolls indicates total illiteracy where writing was concerned.¹¹⁶ Moreover, it should be noted that the shortage of descriptions of social status for most of the oath-takers further compromises the value of the oath rolls in reconstructing literacy rates. As noted in the discussion of freeholders lists above, local elites appear to have been over-represented amongst the oath-takers compared to society as a whole. If it is assumed that there was a correlation between social status and literacy levels, then this further compromises the value of the oath rolls as an indicator of literacy rates in early Hanoverian Devon. Further research is needed on individual parish communities to determine the full social range of the subscribers, in order to make statistical adjustments to the overall literacy levels extrapolated from the oath rolls.

Despite all of these qualifications it is possible to draw conclusions regarding the minimum levels of literacy among those who took the oaths. These figures do bear some comparison with existing research on literacy levels in pre-industrial England.¹¹⁷ At the most basic level, 42% of those who swore the oaths were able to sign their name. Along gendered lines, this breaks down to 22.7% of women who signed and 50.1% of men. These figures compare to national estimates suggesting that at the time of the accession of George I 55% of men were unable to sign compared to 75% of women. This, in turn, represented a substantial advance from the civil war years when 70% of men were illiterate and 90% of women.¹¹⁸ Based on the ability to sign of subscribers to the Protestation Oath 72% of men living in Devon during the 1640s were illiterate.¹¹⁹ Breaking the 1723 figures down by month of swearing yields the figures presented in Table 1 (below).

Month	% Women signed	% Men signed
August	25.0% (688)	62.9% (2,561)
September	22.2% (1,501)	54.3% (5,758)
October	21.6% (1,187)	45.6% (3,375)
November	28.7% (1,581)	45.4% (2,553)
December	18.1% (1,783)	34.1% (1,696)

Table 1. Literacy levels, by month of swearing. Figure in brackets is total sample size.

The decline in literacy levels apparent between the August and December sessions is probably caused by the practice of initialling in lieu of a signature. Alternatively, it may indicate that the August and September oath-takers were of a higher social status than those who swore during the later months. If we assume that the former is true, then it is possible to calculate revised literacy levels based on the returns for August and part of September.¹²⁰ This exercise yields figures of 63.2% literacy for men and 25.4% for women.¹²¹

It is also possible to calculate literacy rates by parish of residence, although because the likelihood of individuals signing their own names seems to have varied according to time of swearing these should be treated with caution. For example, it would not be wise to attempt to map the geographical spread of literacy across the county without employing some method of statistical adjustment to the data from the later oath rolls. However, it is possible to identify parishes where there is evidence of particularly high literacy levels. Here, in order to concentrate on results that might be considered statistically significant only parishes providing in excess of 50 male and 40 female subscribers have been considered.¹²² Table 2 lists the literacy rates of parishes within this category with male literacy levels at 60% or above and Table 3 gives the literacy rates of parishes with female literacy levels at 25% or above.

Parish	Male signed	Sample size	Male literacy (%)
Exeter123	775	860	90.1
Tavistock	128	160	80.0
Stoke Damarel	170	214	79.4
Topsham	105	135	77.8
Plymstock	56	72	77.8
Lifton	49	67	73.1
Topsham	105	135	77.8
Modbury	92	126	73.0
Moretonhampstead	107	148	72.3
Kingsbridge	57	79	72.2
Honiton	72	107	67.3
Hartland	51	77	66.2
St Thomas (Exeter)	82	124	66.1
Bere Ferrers	35	54	64.8
Holsworthy	57	88	64.8
Ashburton	95	148	64.2
Wolborough	45	71	63.4
Plympton St. Mary	41	65	63.1
Bampton	56	89	62.9
Colyton	71	113	62.8
Kentisbeare	32	51	62.7
Hatherleigh	47	75	62.7
Ugborough	57	91	62.6
Northam	68	109	62.4
Sampford Courtenay	35	57	61.4
North Petherwin	33	54	61.1
Tawstock	47	77	61.0
Staverton	43	71	60.6
Heavitree	32	54	60.4

Table 2: Parishes with over male 50 subscribers and male literacy above 60%.

Parish	Female signed	Sample size	Female literacy (%)
Exeter	556	735	75.6
Stoke Damerel	58	105	53.2
Modbury	42	79	53.2
Topsham	71	141	50.4
Kingsbridge	23	51	45.1
Crediton	46	116	39.7
St. Thomas (Exeter)	23	60	38.3
Tavistock	16	42	38.1
Ashburton	31	85	36.5
Silverton	15	42	35.7
Cullompton	28	79	35.4
Plymstock	12	35	34.3
Colyton	20	62	32.3
Bampton	17	53	32.1
Honiton	25	78	32.1
Ottery St. Mary	22	75	29.3
Wolborough	14	49	28.6
Holsworthy	11	40	27.5
Hatherleigh	12	44	27.3
Chudleigh	14	56	25.0

Table 3: Parishes with over 40 female subscribers and over 25% female literacy.

By concentrating on parishes with relatively high numbers of subscribers these figures are largely self-selecting. They indicate relatively high levels of literacy within some of the larger towns in the county that were not excluded by virtue of being incorporated. The highest literacy levels can be found in the city of Exeter, with 9 out of 10 male oath-takers able to sign their names, and 3 in 4 women. The high literacy rates for Stoke Dameral and Plymstock suggest that the returns for Plymouth would provide similarly high literacy levels to those for Exeter had they survived. Several larger towns such as Modbury, Moretonhampstead, Ashburton and Honiton feature, as do the parishes of St. Thomas and Heavitree on the fringes of Exeter. These figures can also be compared to those for the town of Great Torrington, whose inhabitants subscribed on a separate roll held at the North Devon Record Office.¹²⁴ Here male illiteracy was confined to 1 in 4 subscribers (25.7%) and a minority of women (45.8%). Beyond these very general observations little more can be deduced concerning levels of literacy from the oath rolls alone. They may provide the basis for further avenues of research, but without additional supporting information what they can tell us regarding the ability of early Georgian Devonians to read and write is limited.¹²⁵ Nonetheless, the respective male and female literacy rates are at least of the same order of magnitude as might be expected around this time, giving further support to the hypothesis that the subscription to the oaths of allegiance was socially extensive.

Continue to next section|Back to previous

---

115. DHC, QS17/2/5/4, oaths sworn at the home of John Brewer, West Teignmouth; QS17/2/6/1, oaths sworn the Castle, Exeter. [back]
116. Nor should inability to write be interpreted as inability to read. See Cressy, Literacy and the Social Order, esp. ch. 2 ‘The acquisition of literacy’. [back]
117. All calculations of literacy rates are based on entries on the oath rolls where either a signature or mark (including clearly legible initials) appears opposite the clerk’s entry for an individual. Where neither a signature or mark is evident, these have not been included in the calculations. [back]
118. Cressy, Literacy and the Social Order, 176. [back]
119. Cressy, Literacy and the Social Order, 74, 194-95 [back]
120. The nature of the ‘marks’ on the oath rolls dating from the early part of September 1723 differs from those later in the month, suggesting that the practice of initialling evolved as the scale of the process became clear. Therefore, the revised literacy levels are based on all oath rolls for August, and those dated up to and including 12 September. [back]
121. Sample size: men = 4829; women = 1310. [back]
122. Excluding those for whom no mark or signature is given. [back]
123. Exeter data extracted from the Exeter Quarter Sessions oath rolls. The entire city is included, as the rolls do not give parish of residence. [back]
124. NDRO, 2558-2/102/1. [back]
125. For a discussion of the types of sources available to the historian of literacy in Tudor and Stuart England see Cressy, Literacy and the Social Order. [back]