Roads of the Military Survey of Scotland, 1747
A brief inspection of Roy’s Highland road network reveals significant differences from today's layout. Not only are there far fewer roads but many of the old roads are not in the current classification; they can be identified but often they are so minor as to be uncategorised. Sometimes a B-road is followed and less frequently an A-road, albeit with frequent off-road diversions reflecting an earlier course; however, much of the old network follows mountain tracks or farm roads which are thereby confirmed to be very ancient routes. Roy may well be describing a legacy from medieval times: the roads firmly follow the easiest routes, taking the river valleys between settlements rather the militarily defensible high ground as is seen with Wade’s roads and Roman roads.
Movement of goods in mid-18th century Scotland was managed as much by unwheeled as by wheeled vehicles; the clear distinction between a horse-road and a carte-road was recognised: wheels needed a much higher quality of road surface. The minutes of turnpike commissioning committees in the 1770s confirm that wheeled travel was a desperate and exhausting business except in very dry weather. The roads were in serious disrepair: Barrow suggests a worse condition even than in medieval times. Yet horse-roads did not preclude transport of goods. Human-back transport was common: wickerwork baskets or creels(cliabh) were harnessed by carrying bands (fettles) across the chest; split saddles and crook saddles permitted horses to carry large curracks on either side; Travois or slide cars were still more common than carts: loads were carried on a pair of poles whose lower ends trailed on the ground behind the draught-animal; these were called kellachs and were sometimes enhanced by semicircular hoops at the trailing-end, or even small wheels (kellach-cairts). In the parish of Kiltarlity in Inverness in 1790 the minister counted 361 sledges alongside 40 small wagons (coups) and 376 carts.
Notes on 18th Century Roads
General Wade had launched his road-building programme from the 1730s onwards: long distance routes, partly new-build but largely upgrading from horse-road to carte-road. Surveyor’s notes on one preliminary plan refers to ’..a low flatt road and dry, may easily be made carte road’ and ‘.. one little step in the wood to be filled up with stone and lime and made smooth, It is easily made cart road.’
Roy gave us no legends for his maps. Six Wade and Caulfield military roads were completed prior to 1750 and three more were under construction while Roy’s survey was taking place; Roy drew these new 'king's roads' with a much heavier emphasis and a double line; they appear to have been very carefully surveyed. There is second group of older roads with a lighter emphasis yet retaining the double line, and a third lesser category, portrayed as a thin faint single line. We can only guess at the specifications implied. There are many settlements which have no connection at all, so we can reasonably presume that the network he describes was not all-inclusive. The faint single line roads seem to be the least accurately surveyed: they have a more topological style, going in rather straighter lines or simpler curves to a given goal, than might seem likely. Our own limited field-survey of some of the minor roads frequently suggests that he map reflects an implausible course on the ground. There may be uncertainty about the very physical basis of some roads of this period: Hindle observes that medieval roads had grown from habitual lines of travel; they were not necessarily thin strips of track with definite boundaries but rather directional rights of way with legal and customary status leading from one village to another. Some of Roy's minor roads may have been of this nature.
Curiously, a number of more northerly roads appear to peter-out before they reach a very obvious junction or goal, thus revealing a failure in mapping rather than a deficiency in the road system; for example, the road NE of Brechin is clearly labelled ‘to Stonehaven,’ but fails to get there!
There was one area of Scotland where roads were conspicuously absent. North of the Great Glen may well have been beyond the reach of Government although it appeared to be well within the reach of the surveyors: it is represented very well on the map and in great detail; there are many settlements. There are very few roads, however. Fife by contrast is extremely well served by roads no doubt reflecting its agricultural importance. Perthshire and Stirling are less well serviced but southern-Argyll has a honeycomb of roads reaching almost every settlement. Could this reflect differences in the specification adopted by different surveying parties?