![Sample our Arts journals, sign in here to start your access, latest two volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/110799/TOC-Subject-Sample-2021-Arts.jpg"})
ABSTRACT
A detailed study was made of the distribution of settlements in England and Wales as shown on the medieval Gough map of Britain and of the accuracy of their placement. No evidence of clustering is found in the choice of settlements in England, which are more uniformly dispersed than expected even for a random distribution. This suggests that whoever constructed the map was aiming for a uniform distribution in order to make it easier to refer to localities not shown. Evidence is found that suggests that the map was constructed from a series of regional ‘surveys’, including maps, which were individually more accurate than the resultant map of England and Wales overall. The strange shape of Wales, and in particular the absence of any indication of Cardigan Bay, is attributed to an especially poor combination of three regional surveys, each of a similar accuracy to those of the regions of England. Lesser distortions of the outline of England can be partly attributed to similar errors in the assembly of the regional surveys. The possible methods of construction of the map are discussed in the light of these results. A number of misidentifications of settlements on the map came to light in the course of these investigations.
La carte médiévale de Gough, sa géographie du peuplement et la représentation imprécise du Pays de Galles
Une étude détaillée a été faite de la distribution des lieux habités en Angleterre et au Pays de Galles, indiqués sur la carte médiévale de Gough de la Grande-Bretagne, et sur la précision de leur emplacement. On n’identifie pas de tendance générale dans le choix des lieux de peuplement en Angleterre : ils sont plus uniformément distribués que l’on ne pourrait s’y attendre, même pour une distribution aléatoire. Ceci indique que quiconque construisit la carte recherchait une distribution uniforme afin qu’il soit plus facile de se référer à des localités non représentées. Preuve est faite que la carte fut construite à partir d’une série d’enquêtes régionales, qui incluaient des cartes, qui prises séparément étaient plus exactes que l’ensemble de la carte d’Angleterre et de Galles qui en résulta. La forme étrange du Pays de Galles et en particulier l’absence de toute indication sur la baie de Cardigan, est attribuée à la médiocre combinaison de trois levés régionaux, chacun d’une précision équivalente à ceux des régions d’Angleterre. De moindres déformations dans les contours de l’Angleterre peuvent être en partie attribuées à des erreurs similaires dans l’assemblage des levés régionaux. Les méthodes possibles de construction de la carte sont discutées à la lumière de ces résultats. Un certain nombre d’identifications erronées de lieux habités de la carte ont été mis en lumière au cours de cette recherche.
Die mittelalterliche Gough Map, ihre Siedlungsgeographie und die ungenaue Darstellung von Wales
Auf der Grundlage einer detaillierten Untersuchung der mittelalterlichen Gough Map von Britannien werden die Verteilung der Siedlungen in England und Wales sowie die Genauigkeit ihrer Positionierungen untersucht. Es konnten keine Anzeichen von Häufungen bei der Auswahl der dargestellten Siedlungen in England festgestellt werden. Diese sind gleichmäßiger verteilt als man es selbst bei einer zufälligen Auswahl erwarten könnte. Das lässt darauf schließen, dass, wer immer die Karte konstruierte, auf eine gleichmäßige Verteilung Wert legte, um die Lage nicht wiedergegebener Örtlichkeiten durch Bezugnahme auf die dargestellten leichter beschreiben zu können. Es konnten Hinweise gefunden werden, die nahe legen, dass die Karte auf der Grundlage mehrerer Regionalaufnahmen entstanden ist. Diese Vorlagen dürften Karten enthalten haben, die jeweils genauer waren als die aus ihnen abgeleitete Karte von England und Wales. Die ungewöhnliche Form von Wales und speziell das Fehlen der Cardigan Bay kann auf eine nicht gelungene Kompilation dreier Regionalaufnahmen zurückgeführt werden. Diese Vorlagen weisen eine ähnliche Genauigkeit auf, wie jene der Regionen Englands. Kleinere Verzerrungen im Umriss Englands können teilweise ähnlichen Irrtümern bei der Auswertung von Regionalaufnahmen zugeschrieben werden. Vor dem Hintergrund dieser Erkenntnisse werden die denkbaren Methoden des Kartenentwurfs diskutiert. Zusätzlich wurden einige Fehlinterpretationen von Siedlungsnamen festgestellt.
El mapa medieval de Gough, sus asentamientos geográficos y la representación inexacta de Gales
La distribución de los asentamientos en Inglaterra y Gales del mapa medieval de Gough y la exactitud de su ubicación han sido estudiados con detalle. No se han encontrado evidencias de agrupaciones en los asentamientos escogidos en Inglaterra, los cuales están dispersos de manera más uniforme de lo esperada, incluso para una distribución aleatoria. Esto sugiere que quien realizó el mapa buscaba una distribución uniforme con el fin de hacer más fácil referirse a localidades no mostradas. Se han encontrado evidencias que sugieren que el mapa fue construido a partir de una serie de ‘mediciones’ regionales, incluyendo mapas, que eran individualmente más precisos en general que el mapa resultante de Inglaterra y Gales. La extraña forma de Gales y, en particular, la falta de indicación de la bahía de Cardigan, se atribuyen a una combinación especialmente pobre de tres mediciones regionales, cada una de una precisión similar a las de las regiones de Inglaterra. Las distorsiones menores del contorno de Inglaterra pueden atribuirse en parte a errores similares en la combinación de mediciones regionales. Los posibles métodos de realización del mapa son analizados a la luz de estos resultados. Una serie de errores de identificación de los asentamientos en el mapa salió a la luz en el curso de estas investigaciones.
Notes
1. The map was described in detail by E. S. J. Parsons, The Map of Great Britain circa A.D. 1360 Known as the Gough Map (Oxford, Bodleian Library and the Royal Geographical Society, 1958). See also Nick Millea, The Gough Map: The Earliest Road Map of Great Britain? (Oxford, Bodleian Library, 2007).
2. Richard Gough, British Topography (London, 1780).
3. The editor informs me, at the time of writing, that an examination of the physical aspects of the document is currently in process and that the outcome of this will be reported by the end of 2015 or soon after. For the problems involved see especially Elizabeth Solopova, ‘The making and re-making of the Gough map of Britain: manuscript evidence and historical context’, Imago Mundi, 64:2 (2012): 155–168; and Thomas M. Smallwood, ‘The making of the Gough map reconsidered: a personal view’, Imago Mundi 64:2 (2012): 169–180. On the date of the map, Smallwood has noted that ‘Neither Dr Solopova nor I believes that the [Gough] map is likely to have been made outside the period from a third of a century before 1400 to a third of a century after it’ (176). For the older suggestion of a prototype of about 1280, see Parsons, The Map of Great Britain (note 1), 2; and Millea, The Gough Map (note 1), 13–15; Keith D. Lilley and Christopher D. Lloyd, with contributions by Bruce M. S. Campbell, ‘Mapping the realm: a new look at the Gough map of Britain (c.1360’), Imago Mundi 61:1 (2009): 1–28; C. D. Lloyd, and K. D. Lilley, ‘Cartographic veracity in medieval mapping: analyzing geographical variation in the Gough map of Great Britain’, Annals of the Association of American Geographers 99:1 (2009): 27–48. See also their Gough map website, www.goughmap.org/.
4. Lilley and Lloyd, ‘Mapping the realm’ (see note 3):, 4; Lloyd, and Lilley, ‘Cartographic veracity in medieval mapping’ (see note 3).
5. Lilley and Lloyd, ‘Mapping the realm’ (see note 3), 4.
6. Ibid., 12–14.
7. The numbers in are based on information provided with the location data on the Gough map website (http://www.goughmap.org/map/), which is largely in agreement with the conclusions of Parsons, The Map of Great Britain (see note 1). Locations of all the settlements listed in data downloaded from that website were used in the statistical analyses of settlement distributions, but only the locations of those considered to be definitely identified were used in the placement-error analyses.
8. In the data referred to in note 7, and therefore in the present work, the point corresponding to each settlement is taken to be the centre of the base of the vignette representing the settlement.
9. R. D. Clarke, ‘An application of the Poisson distribution’, Journal of the Institute of Actuaries 72 (1946): 481.
10. See note 8
11. The 50 per cent was estimated by eye. No other simple method is readily available for the irregular shapes involved and slight inaccuracies do not significantly affect the following arguments.
12. Described in Bernhard Jenny (Oregon State University) and Adrian Weber (2005–2014; http//mapanalyst.org/); Bernhard Jenny, Adrian Weber and Lorenz Hurni, ‘Visualizing the planimetric accuracy of historical maps with MapAnalyst’, Cartographica 42 (2007); 89–94.
13. The coordinates of the settlements on the Gough map are given in the downloaded data (see note 7) in a set of units that are not defined. The value of the unit determined by the present author is 1.39 cm, using the measurements given by Millea, The Gough Map (see note 1), for the length and breadth of the skin on which the map is drawn. This means that the scales determined using the downloaded data as if the units were centimetres should be reduced by this factor, that is, should be multiplied by 0.72 to obtain true scale factors, and this has been done for all the data reported in this article. The value of the multiplier was confirmed in a personal communication from Lilley.
14. Because the scale of the Gough map is approximately 1:1 million, 1 kilometre on the ground is represented by approximately 1 millimetre on the map.
15. The identification was suggested by Parsons, Gough Map (see note 1), 27
16. This identification may not be particularly significant, because the full set of counties for Wales was not established until 1535. Caernarvonshire, Flintshire, Merionethshire, Cardiganshire and Carmarthenshire were, however, established in 1282, and Pembrokeshire and Glamorganshire were also established quite early.
17. The word ‘survey’ is used here and in what follows simply as a short way of saying ‘acquisition of the information required for producing a map of the area in question’ and does not imply any particular method for such acquisition. The word ‘surveyor’ is used for the corresponding acquirer of the information.
18. Julius P. Gilson, Four Maps of Great Britain Designed by Matthew Paris about A. D. 1250 (London, The British Museum, 1928); P. D. A. Harvey, ‘Matthew Paris’s maps of Britain’, in Thirteenth Century England: IV: Proceedings of the Newcastle upon Tyne Conference 1991, ed. P. R. Coss and S. D. Lloyd (Rochester, NY, Boydell Press, 1992): 109–21.
19. All tabulated data were obtained using the ‘Analyse Old Map’ method in MapAnalyst, as were the data shown in . In order to show the results on a modern map, the data shown in were obtained using the ‘Analyse New Map’ method, and the errors shown do not correspond exactly to those that would be obtained by using the ‘Analyse Old Map’ method. The differences are, however, likely to be small, because the difference between the scale factors and between the values of σ for the two methods of analysis are both less than 1 per cent.
20. Lloyd and Lilley, ‘Cartographic veracity in medieval mapping’ (see note 3), 39, fig. 8, show that when the whole of England, Scotland and Wales is analysed by MapAnalyst, there is a ‘whirlpool effect’ in the displacement vectors in southern England. They claim that this effect disappears when England alone is analysed. The present shows that there is a residual effect.
21. Lilley and Lloyd, ‘Mapping the realm’ (see note 3), 9–10.
22. Brian P. Hindle, ‘The road network of medieval England and Wales’, Journal of Historical Geography, 2:3 (1976): 207–21.
23. See for example, Daniel Birkholz, The King’s Two Maps: Cartography and Culture in Thirteenth-Century England (London, Routledge, 2004), 67; Lilley and Lloyd, ‘Mapping the realm’ (note 3), 2; and Millea, The Gough Map (note 1), 13‒15. The exact nature of any such prototype is still unclear.
24. Lilley and Lloyd ‘Mapping the realm’ (see note 3), 18 and note 58.
25. Birkholz, The Kings Two Maps (see note 23)
26. See Ifor M. Evans and Heather Lawrence, Christopher Saxton: Elizabethan Map-maker (Wakefield, Wakefield Historical Publications, and the Holland Press, London, 1979), for Saxton and his atlas. It is not completely clear to what extent Saxton compiled Anglia, his general map of England and Wales, from his regional surveys, which were of counties or groups of counties, and to what extent he used existing small-scale maps of the country (see chapters 3 and 4 of the book, particularly 34 and 41).
27. D. I. Bower, ‘Saxton’s maps of England and Wales: the accuracy of Anglia and Britannia and their relationship to each other and to the county maps’, Imago Mundi 63:2 (2011): 180‒202, where I argue that the maps Anglia and Britannia were produced largely by scaling, rotating and suitably placing the county maps.
28. A glance at the county maps in Saxton’s Atlas of 1579 shows that the boundaries of an individual county are often shown differently on the maps of adjoining counties from the way they are shown on the county map itself.
29. The suggestion that the rivers might have formed the framework for the map was suggested to me by Professor Charles Phythian-Adams (former head of English Local History, University of Leicester).
30. P. D. A. Harvey, ‘Medieval maps to 1500‘, in Historian’s Guide to Early British Maps (London, Royal Historical Society, 1994), 13‒15. I do not find this suggestion convincing. For illustrations of the development of the outline of the British Isles on portolan charts, Michael Andrews, ‘The British Isles in the nautical charts of the XIVth and XVth centuries’, Geographical Journal 86:6 (1926): 474‒81, is still useful.
31. Andrews, ‘The British Isles in the nautical charts’ (see note 30).
32. The map of Italy is British Library, Cotton Roll xiii.44. I am grateful to Catherine Delano-Smith for information about this map (personal communication, 19 January 2015).
33. David Roffe, Domesday: The Inquest and the Book (Oxford, Oxford University Press, 2000), 125ff.
34. Ibid., 61–62. The hundred, the official subdivision for taxation, justice, police, law and military defence, was indispensable for the tightly-governed and outstandingly efficient local administration characteristic of 13th century England from the Domesday inquest on. The circuits used in 1255, and those probably used in the 1274–1275, inquiry have been tabulated by Sandra Raban, A Second Domesday: The Hundred Rolls of 1279‒80 (Oxford, Oxford University Press, 2004), 16 and 23. The 1279–1280 inquiry was conducted on the basis of single counties (13 counties) or pairs of counties (26 counties).
35. Raban, A Second Domesday (see note 34), 61‒62.