West Kennet avenue is probably of Early Bronze Age date
There is a consensus that stone rows were built from the end of the Neolithic, but that most were erected in the Early Bronze Age (EBA) whilst a few stragglers made late appearances in the Middle Bronze Age (MBA). The evidence to support this chronology is far from reliable, is often circumstantial at best, but is generally consistent and it is currently all we have to work with.
Suggested generalised chronology from Burl, A., 1993. Click on image to open a larger version.
The idea is that the rows owe their origin to the avenues leading from Stonehenge and Avebury (Burl, A., 1993, 66-69). The avenue at Stonehenge is formed by a ditch and bank meaning that it has been possible to date it with some precision to 2100 – 2300 BC, whilst the avenues at Avebury (Beckhampton and West Kennet) are thought to date between 2000 and 2600 BC. If one accepts that the origins of the stone rows were indeed to be found at these two places then it would follow that all the rows would be more recent. One immediately runs into the problem of the row at the highest point of Dartmoor at Cut Hill where the stones had already fallen or been pushed over by 2300 BC. It would therefore seem that the inspiration for the rows may not have been so straightforward. There are very few dates available for stone rows and most are derived from association rather than from the row itself. The only exceptions are at Cut Hill (pre-2300 BC) and Ardnacross (1260-810 BC) which together illustrate their potential longevity. Elsewhere the dating evidence is provided by association and is therefore circumstantial at best. At Broomend of Crichie and Callanish adjacent burials might suggest that these rows were being used for funerary purposes between 1900 and 2100 BC. The problem is we cannot be sure that the funerary activity was in anyway associated with the rows. The rows could have been erected much earlier or indeed much later and had nothing to do with the funerary activity. The same is the case at Askham Fell, Lacra NE and Garrywhin where the associated cairns have been dated using pottery to the Early Bronze Age. We do not however have any proof that the cairns and rows were contemporary. There are three possibly scenarios. The cairn and rows were built at the same time in which case the date of the rows is known. The cairn was built first and the rows added later or the rows were built first and the cairn added later. The problem is we have no evidence as to which scenario is the most likely and therefore do not know at what date the rows were erected. They could be earlier than the dated burials, contemporary or indeed much later. What is clear is that several stone rows are associated with Early Bronze Age material and whilst this is suggestive it is by no means conclusive. Excavated dating material is not the only source of information on the dating of the rows and on Dartmoor some helpful stratigraphic details provide an insight into the problem. At Hurston Ridge a very fine double stone row has been slighted by a later enclosure attached to a reave. The enclosure is likely to be more recent than the row and its construction might have rendered the row obsolete.
Later MBA field boundary leading across the double stone row at Hurston Ridge.
Enclosures such as this are considered to belong to the Middle Bronze Age (MBA) and this relationship is therefore powerful evidence to support the idea that rows had become redundant by this time and were therefore built and used earlier than this. It is however worth noting that this enclosure has not actually been dated. A second site where a stone row is considered to be earlier than a Middle Bronze field system is at Yar Tor where three field boundaries appear to overlie the row.
The triple stone row at Yar Tor seems to pre-date MBA coaxial field boundaries.
Elsewhere at Hook Lake a double stone row is partly buried beneath a later MBA enclosure wall and one stone is even incorporated into a roundhouse. Whilst at Assycombe and Shoveldown Row 1 a MBA reave separates the row into two parts. At Leeden Tor the terminal cairn appears to have been incorporated into a reave and the row itself is interrupted by another reave. At West Sharpitor the lower end of the row is adjacent to a reave. These seven sites provide the strongest evidence that at least some stone rows are earlier than Middle Bronze Age in date and this taken with the EBA associations identified elsewhere seems to be solid evidence to support the idea that stone rows were being built in the Early Bronze Age. How reliable is this conclusion? Is it possible that all is not as it appears? The most surprising thing about all seven sites is that the rows survive surprisingly well despite the later re-use. Is it possible that the rows rather than being earlier than the Middle Bronze Age features were either contemporary or more recent? The answer is that we just can’t be sure. In every instance it is possible that the rows were added later and that the rows builders paid no regard to the earlier features, building their rows across abandoned or even possibly still active fields and enclosures. On balance this interpretation seems unlikely but given the obvious regard being paid to these monuments by the MBA inhabitants of the moor it is possible that stone rows continued to be used and perhaps even built on MBA Dartmoor. We can be fairly sure that stone rows were being built elsewhere in the MBA as radio-carbon dates from two sites in Argyll and Isles at Ardnacross and Ballymeanoch suggest that both were erected at that time. A date of 1370-1040 BC from a stone socket close to the spectacular row at Ballymeanoch and 1260-1810 BC from Ardnacross provide evidence that at least some of the rows in that region were being built or at least used at this time. It does not necessarily follow that rows were also being built in other parts of the country but there is compelling albeit circumstantial evidence to suggest that rows were being erected and used on MBA Dartmoor.
Possible chronological context of individual rows. Single rows are shown blue, double rows green and triple/multiple rows orange. Click on image to open a larger version.
The strongest evidence that the stone row tradition continued into the Middle Bronze Age on Dartmoor is provided by the clear relationship between some rows and the MBA landscape. The acceptance that rows may have been built and used at this time would certainly explain why so many of the slighted rows survive so well and perhaps also explain a number of other relationships. At Shell Top SW, Penn Beacon, South and Holne Moor the rows are positioned very close to and precisely parallel with the adjacent reave implying that they were built taking cognisance of them. Whilst not conclusive it is more likely than not that these rows were built after the adjacent reaves. Furthermore at Merrivale, Drizzlecombe and Black Tor (Stanlake) the rows stand intact immediately next to substantial Middle Bronze Age round house settlements. The survival of the rows so close to these settlements implies that they were important to the people who lived here. Either they regarded them with such reverence that they dare not touch them or more likely they were still being used or even built at the same time as the settlements. Further evidence for stone row contemporaneity with the MBA landscape is provided by the rows situated on the periphery of field systems. At Collard Tor,West and East, Cosdon and Shaugh Moor the rows stand beyond MBA fields. At Shoveldown five of the rows respect the MBA field boundaries, with only Row 1 being slighted by a reave. This situation may indicate chronological depth with Row 1 being built first and the others following the establishment of MBA settlements and fields in the area.
Whilst we have fairly strong evidence that some stone rows are likely to have been earlier than the Middle Bronze Age we also have strong circumstantial evidence that the rows continued to be respected and probably also built. Preliminary analysis of the Dartmoor rows has shown that 34 are likely to have been built during the MBA. These are listed in Table 1 together with reasons why a MBA date is suspected.
| NAME | REASON |
| Black Tor (Stanlake) | Adjacent to MBA settlement |
| Black Tor Avon | Adjacent to MBA settlement |
| Burford Down | Adjacent to MBA settlement |
| Collard Tor E | Built just outside MBA field system |
| Collard Tor W | Built just outside MBA field system |
| Cosdon | Built just outside MBA field system |
| Drizzlecombe 1 | Adjacent to MBA settlement |
| Drizzlecombe 4 | Adjacent to MBA settlement |
| Drizzlecombe row 2 | Adjacent to MBA settlement |
| Drizzlecombe row 3 | Adjacent to MBA settlement |
| Hart Tor, North | Adjacent to MBA settlements |
| Hart Tor, South | Adjacent to MBA settlements |
| Holne Moor | Built next to parallel with a MBA reave |
| Joan Ford’s Newtake | Adjacent to MBA settlement |
| Little Links Tor | Adjacent to MBA settlement |
| Merrivale 1 | Adjacent to MBA settlement |
| Merrivale 2 | Adjacent to MBA settlement |
| Merrivale 3 | Adjacent to MBA settlement |
| Merrivale 4 | Adjacent to MBA settlement |
| Merrivale 5 | Adjacent to MBA settlement |
| Merrivale 6 | Adjacent to MBA settlement |
| Penn Beacon S | Built next to parallel with a MBA reave |
| Sharpitor NE | Built next to a MBA reave and adjacent to MBA settlements |
| Shaugh Moor | Built just outside MBA field system |
| Shell Top SW | Built next to parallel with a MBA reave |
| Shoveldown 2 | Close to MBA reave and settlement |
| Shoveldown 3 | Close to MBA reave and settlement |
| Shoveldown 4 | Close to MBA reave and settlement |
| Shoveldown 5 | Close to MBA reave and settlement |
| Shoveldown 6 | Close to MBA reave and settlement |
| Stalldown SE1 | Adjacent to MBA settlement |
| Trowlesworthy 1 | Adjacent to MBA settlement |
| Trowlesworthy 2 | Adjacent to MBA settlement |
Table 1 Dartmoor stone rows with possible MBA origins and or use.
Previously it has been assumed that all the Dartmoor rows belong to the Early Bronze Age and were abandoned once more intensive activity started in the MBA. The alternative idea of the rows continuing to be used and built is not contradicted by the existing evidence and indeed more comfortably fits it. There is now no need to explain why the rows at Merrivale and Drizzlecombe survive so well despite their close proximity to MBA settlements, the survival of stone rows apparently incorporated into MBA boundaries and fields now also makes sense. If one accepts that the rows continued to be used throughout the EBA and MBA then the apparent contradictions are satisfactorily resolved. It may also explain why there are so many rows on Dartmoor compared to other areas. The area was intensively utilised during the MBA in particular and therefore if one accepts that the rows were contemporary with this activity then the reason for the density is self evident.
Having established that the Dartmoor stone rows were probably built and used for around 1600 years from about 2600 BC until 1000 BC it is perhaps worth having a look at how it all started (Table 2). In the Early Bronze Age Dartmoor was a very different place to the one we visit today or even the one at the end of the MBA. There were very few settlements and land-use was of incredibly low intensity. There were a handful of small Neolithic areas of ritual significance each with either a long cairn or substantial round cairns. The earliest row builders as in Northern Scotland appear to have been drawn to these areas and added rows such as at Corringdon Ball, but fresh areas of ritual/funerary activity were established the foremost amongst these being Butterdon Hill where large numbers of cairns and several rows were added presumably over a period of time. Most rows however were built individually a long way from habitation but perhaps close to a developing network of tracks. Conies Down, Higher White Tor and Trendlebere Down represent examples of this type of row. It may be significant that the rows were erected away from home, but close to routes that frequently travelled. These isolated rows never developed into larger scale ritual areas and may have been abandoned at a relatively early date and certainly before the MBA. Certain rows however became the focus for later activity and were expanded by the addition of one or more rows, cairns and sometimes as at Fernworthy by the addition of a stone circle. These ritual complexes including the one at Lakehead Hill were probably used for a relatively long time and may even have continued in use through the MBA. Land-use intensified considerably in the MBA with thousands of new houses, hundreds of enclosures and huge field systems being built over large swathes of the moor. Reflecting this expansion, several fresh ritual complexes were probably established and compared to the earlier ones they were grander and more sophisticated. At Merrivale at least six separate rows probably belong to this period. There may have been an earlier row or two at the site, but it seems probable that more were added as the complex became increasingly important to the people who lived nearby. Compared to the earlier period the row complexes seem to have been built closer to home and this may explain why they become more sophisticated. Perhaps more time was spent at them now that they were closer to habitation. It is almost certainly significant that the three densest clusters of rows those at Shoveldown, Merrivale and Drizzlecombe are situated adjacent to MBA settlements. This relationship provides evidence of a link between the MBA and the rows. Many of the isolated rows built at this time were apparently positioned on the edge of fields, beside reaves and sometimes a short distance from the settlements. Many rows probably continued in use until Dartmoor was de-populated at the end of the MBA, although in all likelihood some had been abandoned.
| NAME | REASON |
| Assycombe | MBA reave built across the row |
| Brent Fore Hill | Close to and visual links to Neolithic long cairn |
| Butterbrook 1 | Within the Butterdon Hill ritual and funerary complex |
| Butterbrook 2 | Within the Butterdon Hill ritual and funerary complex |
| Butterdon E. | Within the Butterdon Hill ritual and funerary complex |
| Butterdon Hill Row | Within the Butterdon Hill ritual and funerary complex |
| Cantrell | Within the Butterdon Hill ritual and funerary complex |
| Challacombe Down | No visible MBA archaeology in vicinity |
| Cholwichtown stone row | No visible MBA archaeology in vicinity |
| Conies Down | No visible MBA archaeology in vicinity |
| Corringdon Ball, North | Close to and visual links to Neolithic long cairn |
| Corringdon Ball, South | Close to and visual links to Neolithic long cairn |
| Cut Hill | No visible MBA archaeology in vicinity |
| Fernworthy 1 | EBA beaker from nearby cairn |
| Fernworthy 2 | EBA beaker from nearby cairn |
| Fernworthy 3 | EBA beaker from nearby cairn |
| Glasscombe Ball N | Within the Butterdon Hill ritual and funerary complex |
| Glasscombe Corner | Within the Butterdon Hill ritual and funerary complex |
| Hamel Down | No visible MBA archaeology in vicinity |
| Higher White Tor | No visible MBA archaeology in vicinity |
| Hingston Hill | Forms focus of ritual area |
| Holne Ridge North | Forms part of a ritual area |
| Hook Lake | Partly underlies a MBA enclosure |
| Hurston Ridge | MBA boundary built across the row |
| Lakehead Hill E | Forms part of a ritual area |
| Lakehead Hill summit | Forms part of a ritual area |
| Langstone Moor | No visible MBA archaeology in vicinity |
| Laughter Tor 1 | No visible MBA archaeology in vicinity |
| Laughter Tor 2 | No visible MBA archaeology in vicinity |
| Leeden Tor | MBA boundary built across the row |
| Penn Beacon SW | Associated with a substantial cairn |
| Piles Hill | Within the Butterdon Hill ritual and funerary complex |
| Ringmoor Down | No visible MBA archaeology and within funerary complex |
| Sharpitor NW 1 | No visible MBA archaeology and within funerary complex |
| Sharpitor NW 2 | No visible MBA archaeology and within funerary complex |
| Sharpitor W | MBA boundary built across the row |
| Sherberton Common | No visible MBA archaeology in vicinity |
| Sherberton Row | No visible MBA archaeology in vicinity and link with stone circle |
| Shoveldown 1 | MBA boundary built across the row |
| Soussons Down | No visible MBA archaeology in vicinity |
| Spurrell’s Cross | Within the Butterdon Hill ritual and funerary complex |
| Stalldown Row | No visible MBA archaeology and within funerary complex |
| Tottiford Reservoir I | Forms part of a ritual area |
| Tottiford Reservoir II | Forms part of a ritual area |
| Treeland Brake | Close to and probable visual links to Neolithic long cairn |
| Trendlebere Down | No visible MBA archaeology in vicinity |
| Upper Erme Row | Forms focus of ritual and funerary complex |
| White Ridge | Forms focus of ritual and funerary complex |
| Yar Tor | MBA boundaries built across the row |
| Yardworthy | No visible MBA archaeology in vicinity |
| Yellowmead Down | Associated with a concentric stone circle |
Table 2 Dartmoor stone rows with possible EBA origins and or use
The remarkable survival of a rich prehistoric landscape on Dartmoor provides the information to permit an informed discussion of relative dates. Elsewhere this is not the case with the exception of Bodmin Moor where the rows at Leskernick, Craddock Moor and Carneglos are clearly associated with MBA settlements whilst those at Minions, Stannon, East Moor and Trehudreth Downs form parts of ritual complexes. It would seem therefore that on Bodmin Moor rows were also built and used throughout the EBA and MBA. On Exmoor we do not have visible evidence for any associated field systems or settlements but a date of 2460 – 2190 BC from a cairn next to Lanacombe 2 together with a date of 1560 – 1430 BC from a gully adjacent to Lanacombe 3 and 2040 – 1880 from a buried soil at the same site supports the idea that the rows on Exmoor were built in the EBA and used into the MBA. Clearly further work is needed to confirm this but these results are consistent with the dates suggested for the rows elsewhere in South West England.
Moving across the Bristol Channel to Wales dating of the rows becomes much more difficult. The rows survive in the same sorts of location and share similar types of association, but the dating evidence is lacking. A couple of collared urns in the kerbed cairn at the top of the Fonllech row is about as good as it gets. This makes any discussion of dating fruitless beyond noting that it is likely that as in South West England the rows made their appearance at the very start of the EBA and probably continued to be built and used until the end of the MBA.
In northern England at Askham Fell and Lacra NE pottery recovered from terminal features suggests an EBA date. Both rows form part of a ritual complex similar in character to those identified on Dartmoor, but there is no evidence to indicate whether they continued in use into the MBA. The distinctive megalithic short rows of Argyll and Isles and Central Scotland are very different in character to those further to the south. Again there is limited circumstantial dating evidence, but what exists suggest that these stones were again erected from the EBA until the MBA. At Balnaguard a beaker and food vessel found at different times close to the row provide strong circumstantial evidence that the row formed a focus for funerary activity in the EBA. At Ardnacross C14 dating suggests activity around the row between 1260 and 810 BC and at Ballymeanoch from 1370 to 1040 BC. These dates suggest that rows in Argyll and Isles were being erected in the MBA and whilst it is possible that all the rows in this region belong to this period some maybe earlier. A cluster of EBA burials near to the southern end of the once impressive avenue at Broomend of Crichie in Aberdeenshire provides further evidence that the rows were probably built around this time. In the Western Isles dating of the rows at Callanish relies on associated beakers which Burl has suggested might mean that the site was active around 2000 BC although it may have been erected quite some time before.
Finally in Northern Scotland the evidence, both artefactual and contextual, suggests that the rows belong to the EBA. Beakers from associated cairns at Battle Moss and Garrywhin provide strong albeit circumstantial evidence that the rows were being used at this time. Perhaps more convincing is the recurring association with Neolithic chambered cairns. In particular, the cluster of rows around Loch Watenan is associated with several chambered cairns most of which have specific visual links to the rows. This in itself might suggest a Neolithic date for many of the rows, but whilst this cannot be discounted it is more likely that the rows represent a continuation of a ritual theme.
Conclusion
Cumulative, but largely circumstantial evidence suggests that the stone rows in Great Britain were probably built from the very Early Bronze Age until the end of the Middle Bronze Age. Most are likely to have been built and used between 2600 and 1900 BC, although some especially on Dartmoor, Bodmin Moor and Argyll and Isles were being built well into the MBA. The rows in Northern Scotland may have been built in the Late Neolithic and abandoned before the MBA, whilst most of the rows in Argyll and the Isles may have been built in the MBA.
References
Burl, A., 1993, From Carnac to Callanish – The Prehistoric stone rows and avenues of Britain, Ireland and Brittany, Yale University Press.
First Published: 13th December 2019
Last Updated: 13th December 2019
The Stone Rows of Great Britain 