Archaeology Tuesdays – Lines in the Landscape

In recent weeks, I’ve talked about using archaeology to look at landscapes, either through amassing data from different excavations or through using a museum object as a springboard to look at the bigger picture. Sometimes, though, a single feature can run for long distances, not respecting modern boundaries. Things like railways and Roman roads are good examples of things that cut across the land, more often than not starting and ending outside North Hertfordshire. Ley lines are not a good example, as they don’t exist (perhaps a subject for a future post).

There is a ditch and bank that runs across Therfield Heath for more than 1.4 km. Starting in a valley east of New Road (which runs between Baldock Road and Therfield village), it climbs the hill in a northeasterly direction before turning to a more easterly alignment at the edge of the scarp. It runs north of the barrow cemetery for 390 m before turning to the southeast, where it descends the hill towards Briary Lane and is lost in the grounds of the former Wicker Hall. Although it is undated, it is earlier than Napoleonic ploughing and some of the trackways on the heath; its relationship to the Bronze Age barrows suggests that they were already there when it was dug. In the valley bottoms, it is hidden by a layer of colluvium (soil washed downhill). All told, it is something of a mystery.

Yet more mysterious is an even longer feature, visible on aerial photographs (and Google Earth™, of course!). It runs for almost 3.5 km from near Duckpuddle Bush on New Road, north of Therfield, to just north of Hatchpen, east of the A10. Both ends of the feature simply fade out, and it was probably longer in both directions. It bears no relation to field or parish boundaries, the topography or a pre-modern field system visible on Lidar. Along its course, it crosses five low ridges and the slight valleys between them, on the southern edge of a dry valley that runs east to join the northeast flowing stream in Wardington Bottom.

On the satellite view, it crosses another long line in the landscape, which shows as an angled white stripe. This is a twentieth-century pipeline of some sort, and its whiteness comes from the chalk turned up by digging the trench. The feature we are looking at isn’t one of these, as it shows as a narrow band of generally darker material, showing that the feature results from many years of silting rather than piling the excavated material back into it and perhaps that it was not dug into the underlying bedrock.

Part of it may appear in the Hertfordshire Historic Environment Record as monument number 17018. The record states: ‘A linear cropmark with a central dark toned area; logically a ditch, with lighter-toned banks on either side. It is likely to be archaeological rather than natural, perhaps a ditched track or a boundary.’ The evidence cited is a vertical aerial photograph dated 2010. It makes little sense as a track, as its runs against the grain of the landscape.

Something as long as this ditch is best described as a boundary. The aerial photographs suggest that it was little more than 2 m wide at the top, so we are not looking at something the size of Offa’s Dyke. Instead, we are probably looking at a more localised boundary, albeit on a considerable scale and part of a landscape rather than a single settlement or farmstead.

The earliest such boundaries in Britain date from the early Neolithic: the oldest dated example is on the west side of Hambledon Hill in Dorset, where radiocarbon determinations suggest that it was built about 3600 BC. Here, the ditch ran along the crest of the hill for almost 3 km, nearly as long as that in North Hertfordshire. However, excavation showed that at Hambledon Hill, the ditch runs in short segments, like the causewayed enclosure associated with it. Our local example appears to have been dug as a continuous feature.

Long boundary ditches became more widespread from the Middle Bronze Age on, after about 1500 BC. This was a time of population growth and increasing clearance of the land for agriculture, and such features are sometimes called ‘ranch boundaries’. At the same time, the development of bronze weapons since before 2000 BC had fuelled the rise of a warrior élite, the people whose families were often buried in the round barrows that dot the landscape. As the Bronze Age went on, so these families began to compete (in other words, fight) with each other for control of the land, and some of the linear boundaries built at this time may be early territorial markers.

On the opposite side of the valley from the eastern end of the ditch is Whiteley Hill, the site of a Late Bronze Age defended settlement. Jack Wilkerson and Mary Cra’ster excavated several trenches across its concentric ditches in 1957. We have met the pair on a previous #ArchaeologyTuesday, where we learnt about their work at Aldwick in Barley. At Whiteley Hill, they also excavated a pit located between the two settlement ditches and found most of two smashed storage jars. Although they dated the material to the Early Iron Age (about 850-400 BC), a reassessment of their work by Steward Bryant in the 1980s showed that the pottery is a Late Bronze Age type.

The entrance to Whiteley Hill is on the north side, facing away from the linear boundary ditch. This may be an indication that the focus of the settlement and its farmland lay more toward Burloes than in the valley to the south that the ditch follows. Trial trenching south of Reed Hall in the 2010s uncovered traces of Late Bronze Age settlement, but this is a site overlooking a valley further south, in the drainage basin of the River Quin, a tributary of the Lea. Perhaps we should be looking for Bronze Age activity on the top of the ridge west of Therfield. This is an area where several ring ditches show where earlier Bronze Age burial mounds once stood: Whiteley Hill sits among a group of three. There are also cropmarks of a cluster of ditches that look very like a settlement just below the crest on the north side of the ridge, northwest of Wimsey Hall Plantation. Could this be a farm in the land the boundary marks off?

Might we be looking at a boundary ditch later than the Bronze Age? This seems less likely for several reasons. During the Middle and Late Iron Ages, boundaries often took the form of multiple diches (like the Mile Ditches in the north end of Therfield parish, extending north into Cambridgeshire) or pit alignments (like one that runs for more than 4 km across the north of Iron Age Baldock). The way this one curves across the landscape is more typical of Bronze Age types, and its scale is too small for one of the early medieval dykes that are characteristic of south Cambridgeshire and elsewhere. The way it ignores the Roman road Ermine Street (now the A10) is also an indication that it is likely to be older.

One interesting feature of this long ditch is that its shape mirrors that of the ditch at Therfield Heath. Both have an eastern part trending about 25° south of east (a bearing of about 115°) and a roughly west-east northern section; as the western part of the long boundary has not (yet) been traced, we cannot say if this followed the same approximate bearing as the smaller. On the other hand, the westernmost part of the boundary ditch is pointing towards Thrift Hill, and it may be that this was its intended destination. Thrift Hill – which deserved to be the subject of another post – was a focus of enigmatic activity for millennia, from the Neolithic until recent centuries. As a focal point in the scarp of the hills, this would hardly be surprising.

The question is: why do the two ditches mirror each other but at different scales? Neither is forced to follow its course by the topography; indeed, they seem almost to work against it in places. Do they perhaps define a territory between them, focused on the dry valley? Here, the difficulty is that the long boundary ditch lies towards the foot of the slope on its southern side, while the shorter lies some distance beyond the crest on its northern side. It is probably best to conclude that their similar alignments are nothing more than coincidence and that while the long one is a territorial boundary, the shorter is more likely related to the barrows on Therfield Heath.

We are thus left with a lot of unanswered questions. This is not unusual in archaeology and can drive further research in the hope of finding some answers or, at least, pointers in the right direction. The recognition of this long boundary was unexpected, so we are at the start of a journey of exploration.

Written by Keith Fitzpatrick-Matthews