Roman Roads, A Simple Surveying Technique.
For many years I travelled the south of England in my work as a consulting engineer often on roads, that used as their base, the roman road system. At one time my base was in Chichester and I would often take the opportunity to walk on parts of the Stane Street. I, like many others, admired the skills of the Roman Engineers in laying out their roads and often wondered how they managed to layout such straight roads and know the directions from one town to another. My interest was recently re-kindled when I watched the TV programme with Adam Hart-Davis, “What the Romans Did For Us”, which suggested a technique using the groma’ assumed by many writers as the method used by the Roman surveyors. The drawback, as Adam demonstrated, was that it would have been extremely slow and tedious to use.
I thought again about how the Roman surveyors could have been able to get their bearings with such accuracy. Here I propose a method which would have been quick, easy, and well within their capabilities. I am not a historian so no doubt many who have professionally studied Roman roads may want to take task with my assumptions, however, all writers seem to agree that there is no record of how the Romans surveyors laid out their roads so there is no reason to suppose that they did not use this method. The proof would be to try this method in the field and see what problems were met.
We know the Romans could make accurate measurements of distance across country by witness of their mile posts on their roads. They also used an Hodometer to measure distances. This was a wheeled instrument that dropped pebbles into a container to count the revolutions of a large wheel. In the planning of the roads the Hodometer would almost certainly not have been used, as measuring distances along an undulating course would not have given an accurate determination of the straight line distance between two points. A surveyor’s method would be required. The Romans had knowledge of the Greek’s surveying instrument called the dioptra that could measure angles. This was a sophisticated instrument invented by Hero of Alexandria and similar to our modern theodolite. It is described by M J T Lewis in his book, Surveying Instruments of Greece and Rome. Also an excellent overview of Roman surveying methods can be found at:
Dr. Lewis, however, has kindly pointed out to me that there is no evidence whatever – whether in the Corpus Agrimensorum, which is a large collection of texts specifically on surveying, or anywhere else, that the dioptra was ever used in the western empire. To my mind the fact that we do not have any written record of how the Romans laid out their roads would not preclude the undocumented use of the dioptra in laying out their roads. On the assumption that they did not use the dioptra all they would need is the method of duplicating the angle between two sighting lines. For this they could use either rods to mark the sight lines or a simple plane table. They would also need some accurate means of measuring the distance between two points from which they would take their sight lines. For this they would use rods of a known length as described by various writers.
The Romans would have been able to locate due North or due South with some accuracy. My first thoughts were that they used Polaris, the pole star, however in my rush to get my ideas down on paper I forgot that Polaris would not have been suitable to determine true north as it was not in the position it is today due to the movement of the earth’s axis – the precession of the equinoxes] Dr. Lewis kindly corrected me. The Romans presumably knew the stellar methods of the Egyptians to determine true north and we know that sun dials were very popular in Roman times so they could easily locate due south. It would therefore have been easy for them to a north-south line by use of a ‘gnomon’, a simple pole placed to determine when the shadow of the sun is shortest, i.e. at midday when it is due south. They also used the ‘groma‘ to survey and determine accurate right angles.
Let’s plan the road from Chichester to London, NOVIOMAGVS to LONDINIVM known by us as Stane Street.
For the sake of this illustration I will assume just 3 sites between the two towns. From their general knowledge of the country side the Roman surveyors would have known from their army and scouting parties the general direction that the road would have had to follow and of any prominent features along the route. e.g. hill forts, burial mounds and features like Stonehenge.
Let’s start from Chichester. As he looked north towards the hills of the south downs, the surveyor would have known the approximate point on the hills that the soldiers would have taken as they descended down to Chichester. If there was a natural feature he would note it, or he would arrange with his helpers to start a fire at some convenient point on the hill on the days he wanted to survey. If the surveying was to be carried out on a calm day, then a smoky fire would have been useful, at night a brazier or beacon would be lit.
He lays out his north south line using two poles. At some point on that line he then sets up his plane table or uses another pole to sight to his marker/feature on the hill. He can now determine the angle between due north and his feature/fire.
Using his gromatici, [rod men], to measure the distance, the surveyor sets up at another point due west or east, a known distance from the first where he again lays out a north-south line and determines the angle to the feature/fire on the hill. Note he doesn’t need to know the actual measurement of the angles, he just has to be able to reproduce them. He now has a base line of known distance and two angles.
There will be a number of good sighting locations at the top of the hills between London and Chichester. For instance from the top of Bignor Hill.
He repeats this process at each of his features/fire sites. He can now reproduce the measurements using a scaled base line to produce similar triangles. Note the scaling can be carried out using a large surface. Typically we would think this would be done on paper but some writers suggest the surveyors would use a large floor with moveable objects to mark the observation points and the various obstacles. For greater accuracy the plot can be made in a large field. Alignment with due North would be required.
He will then get a result like this, when he can then produce a straight line to join the two towns and determine the angle from due north or south that he needs to plot his road from Chichester to London.
In the case of Stane Street, the road starts from the south end of London Bridge apparently directly aligned on Chichester’s East Gate, however the Roman surveyors diverted the road at various points to take into account the natural obstacles and the suitability of the ground to support a road, and the actual layout of the road is approximately:
A good account of the needs to vary the route of the road can be found in the book:
“An illustrated history of Roman roads in Britain by David E. Johnston”
copyright Adrian Kerton 2005