Chapman’s proposal for the Nottinghamshire map was issued at the end of May 1774 and he hoped ‘that the proposed Number of Subscriptions will be raised soon enough to undertake the Survey this Summer;’. The survey was undertaken, because the map was published in May 1776, however Chapman did not leave a record of how it was conducted, who was involved or how long it took. For any insight we must turn to the evidence from contemporary surveys of other counties. In general these new surveys were conducted to an unprecedented level of accuracy and topographic detail. How does Chapman’s map compare?
The Field Survey
Whereas the majority of 18th century county maps had been compiled from ‘the best authorities’ in the workshops of the London map publishers, the new maps were based on a field survey. Unlike earlier surveys which relied on the road traverse to establish positions and distances, the majority of the county surveyors used precision instruments to carry out a careful triangulation of the county. The field survey was undertaken by a small team comprising the surveyor, possibly an assistant, some labourers to handle the survey chain and poles, and horses to transport all the instruments and baggage.
In addition the team had the help of local ‘guides’ with knowledge of the landscape. The team would traverse the county recording all the necessary information on draft plans and in the surveyor’s notebooks. The time taken for the survey depended on the area to be surveyed, the scale of the map and the manpower available.
Two summer seasons was typically enough time to complete the survey, with two or more surveyors involved for the larger counties. Henstock(2003) writes that the map ‘was surveyed and produced in 1774 although not published until 1776’. It is more likely that Chapman’s survey took place in the summer months of 1774 and 1775. We know that other county surveyors had assistants, Armstrong in Northumberland, Yates in Lancashire and Burdett in Derbyshire. It would be reasonable to assume that Chapman too, employed an assistant.
The premium offered by the Royal Society in 1759 was intended to stimulate the production of good maps that would give
‘A Complete Knowledge of the Situations, Bearings, Levels and other Topographical Circumstances of this Kingdom, being of great use in planning any scheme for the Improvement of Highways, making Rivers Navigable and providing other means for the Ease and Advancement of National Commerce’ (Harley 1964a p.44).
To this end the Society wished to encourage ‘proper surveyors’ to make ‘an Accurate Actual Survey of any county’, using the theodolite, the plane-table and the perambulator. This set a standard for surveys which was generally adhered to and many surveyors included ‘triangulation diagrams’ on their maps in order, as Yates wrote, ‘to convey an Idea of the Labour and Precision with which this Survey has been carried on and completed’ (Harley 1964a p.110).
Chapman was one of those ‘proper surveyors’ familiar with scientifically based survey techniques through his association with John Rocque (Appendix 3). Chapman had assisted Burdett in Derbyshire and had published Armstrong’s map of Durham in 1768, both of which included a triangulation diagram. It is reasonable to assume that Chapman’s surveys were based on triangulation, but none of his maps include a diagram to confirm this. However two observations by the antiquarian Richard Gough provide circumstantial evidence of the accuracy of Chapman’s surveys (Gough 1780):
‘Surveys on large scales were reserved for the labours of a Rocque, a Jefferys, and a Taylor, and a Chapman. … I invert the chronological order for the climax of merit’ (p.xvi)
‘John Chapman, land-surveyor, began in 1772 for Mrs. Rocque, a map of this county [Essex] … This is one of the most accurate of our county maps’ (p.369).
Triangulation began with the accurate measurement of a base line using the chain, a flat stretch of country was ideal for this purpose. Yates laid out two baselines, of 10 miles and 6 miles, for his survey of Lancashire, ‘carefully measured on the Sea Beach’ (Harley 1964b p.113) and William Beighton used straight sections of the Roman roads in Warwickshire (Macnair 2010). A theodolite fitted with a telescope was used to make angular measurements from each end of the line to selected man-made and natural landmarks, and from these to other landmarks creating a triangular framework of ‘principal stations’. Within each triangle of the survey other features were located in a similar fashion.
Although we know nothing about Chapman’s triangulation of Nottinghamshire, it would appear that one was produced during the abortive Jefferys’ survey. A catalogue of his ‘Drawings & Engraved Maps,…’ published in 1775 (British Library Cartographic Items 203.dd.7) includes the following entry, undated and attributed to an unknown surveyor ‘CM’:
‘Series of angles relating to Survey of Nottinghamshire’.
We can only speculate whether Chapman had access to any of Jefferys’ working papers and in the absence of any surviving documents, can only conjecture what principal stations Jefferys and Chapman might have used. A good candidate for the baseline might have been the old Roman Road, the Fosse Way. A straight section between Syerston and Cotgrave is a possibility. The principal stations were generally high ground and church towers although it would appear that natural landmarks could even include a ‘clump of trees’. Yates used 20 in Staffordshire and 30 in Lancashire, and Burdett used 18 stations in Derbyshire. In Leicestershire, a similar area to Nottinghamshire, Whyman used 19. The average line of sight between stations was 8.3 miles in Leicestershire and 11.9 in Staffordshire. Bearing this information in mind, and with the evidence of Chapman’s map, we can conjecture a triangulation for Nottinghamshire with a base line on the Fosse Way and 18 stations, including out of county landmarks such as Belvoir Castle and Lincoln Cathedral.
Excluding the Lincoln station, the average line of sight would have been 10.6 miles. The Ordnance Survey triangulation of England and Wales, conducted by William Mudge in 1809, positioned two of its principal stations in Nottinghamshire on the conjectured stations at ‘Gringley Beacon’ and ‘Sutton Hill’. It also used Lincoln Minster.
Latitude and Longitude
The county surveyors displayed further evidence of scientific accuracy by their recording of latitude and longitude (Laxton 1976). In order to assess the positional accuracy of Chapman’s map the co-ordinates of 11 locations in the county were compared with those taken from the map.
On average the difference in latitude is negligible at only -0.06%, most likely a systematic measurement error. In respect to longitude all the errors are positive with an average of 2.72%. Is this explained by the choice of meridian? In the later decades of the 18th century the prime meridian of Greenwich was commonly employed, but earlier in the century local meridians had been used as well as St Paul’s Cathedral in London. Chapman’s map declares that the degrees of longitude are ‘west of London’ but there is no information as to his choice of meridian. The longitude of St Paul’s is 00 05’ 53’’ W, suggesting that Chapman’s ‘meridian’ was somewhere east of the Cathedral in the vicinity of Rotherhithe, which is an unlikely choice. A more reasonable assumption is that Greenwich was the meridian and the differences are due to a combination of shrinkage, while the wet printed maps were drying, and systematic measurement error. Shrinkage was a recognized problem and some allowance was made for this by the engraver. Whatever the reasons, the error was not excessive and not unusual: Burdett calculated that his meridian through Derby Cathedral was 10 27’ west of London, an error of 1.75%; Yates placed St Mary’s Church in Stafford 20 3’ west, an error of 3.19%; and Prior’s map of Leicestershire has St Martin’s Church, Leicester, at 10 5’ west, an error of 5.35%.
The scale of the map determined the level of topographical detail surveyed within the triangulation. The 2 inch maps included such fine detail as field boundaries, although Laxton(1976) questions reliance on their accuracy.
They are absent from 1 inch maps, but nevertheless Chapman’s map is still rich in topographical detail (Henstock 2003).
In common with all the other new county surveys, the map is a break with the past, a modernising of earlier representations of the Nottinghamshire landscape. However it is important to recognise that the topographical details are what Chapman chose to include. The decisions he made will have been influenced by his previous experience as a surveyor, the standards set by others and the expectations of those for whom the map was intended.
Roads are clearly marked: turnpikes with their mileposts and toll-bars; and narrow bridle roads. Another symbol of the maps modernity is the Chesterfield Canal, although not fully opened until 1777, the route is shown snaking across the north of the county. Settlements are presented in more realistic detail with buildings shown in plan although churches are still in elevation. Woodland and forest is still represented by groupings of stylised trees although the area of each is more precisely mapped than previously. Tree symbols are used to show avenues on some of the parklands and estates are still enclosed by stylised paling, but presumably more precisely located, and buildings are shown in plan. An innovation is the inclusion of commercial activities. The approximate of 90 windmills is shown by symbols in elevation, the majority to the east of the county. Some 40 watermills are represented by small wheels located on many of the county’s rivers, although not the Trent. Coal pits are shown to the north and west of Nottingham, represented by small circles. Particular buildings, shown in plan, are named as forges, engines, a bleach yard, a paper mill, and a brick kiln. As well as settlements and enclosed estates and areas of woodland, the map shows many areas of common grazing land. A large expanse of open forest land, Sherwood Forest, stretches northwards from the outskirts of Nottingham to Mansfield and Retford. In addition there are other pockets throughout the county named common, moor, warren, wold, carr, smeethe, and meadow, or simply left unnamed. Over the next 50 years the Enclosure Acts would see much of this common land enclosed. Finally Chapman included roads, settlements and estates beyond the boundary of the county.
Assessing the reliability of the map as an accurate representation of the late 18th century landscape is a difficult exercise (Laxton 1976) but a necessary one in order to evaluate properly the historical information contained in the map. The approach would be to make comparisons with other material, for example estate maps and written descriptions of the topography. A pointer to further research.
In the absence of any contemporary evidence we can only speculate on the conduct of Chapman’s survey and draw parallels from what has been written about other counties. The shape of the triangulation is, at best, an informed guess, but the positional accuracy of the map supports the argument that Chapman’s survey was based on a geodetic survey. The map is evidence of what features in the landscape Chapman surveyed, or at least what he chose to represent on his map, and the topography is richly detailed. Chapman completed the survey expeditiously, perhaps because he had the benefit of the earlier work by Jefferys. In any event it is likely that by the winter of 1775 he had completed the manuscript map and was ready to start engraving. It is this artistic stage in the life-story that we turn to next.