Diggers standing by each of the pits during excavation 2005

Diggers stand by each of the pits during the 2005  Scottish National Trust excavation, giving an idea of the scale of the monument at Warren Field. Photo: Murray Archaeological Services Ltd

‘I’ll need it by the end of the week’ is a stock workplace demand. As it is issued you can calculate, with all due aggrievance, exactly how much time this leaves to complete the task. But just how long have we been able to model future time? And when did we start on the path that led us to that bane of modern life: the deadline? Recent research by Vince Gaffney, Professor of Landscape Archaeology and Geomatics at the University of Birmingham, and a group of researchers indicates that the answer may lie in Aberdeenshire. They believe that the dawn of timekeeping stretches back far further than the emergence of the earliest formal calendars in around 3,000 BC.

bone plaque

This Upper Palaeolithic bone plaque from Abri Blanchard, France, was crafted around 30,000 BC. The engraved sequence of circles is believed to represent the waxing and waning moon. Photo: Emilia Pasztor

In the 1960s and 1970s the scholar Alexander Marshack suggested that markings on a handful of Upper Palaeolithic objects, such as the Abri Blanchard bone plate from the Dordogne region in France, represent a waxing and waning moon. Marshack argued that these carvings, manufactured around 30,000 BC, could have served as crude calendars. Current consensus, however, is that these engravings are simply tallies, recording observations of lunar events as they unfolded, rather than making predictions about the future. When clear evidence for formal lunar calendars does appear in the 3rd millennium BC, it comes from Babylonia, that intellectual power house in the cradle of civilisation. Is it really possible that elementary timekeeping had been mastered thousands of years earlier in backward Britain?

Time trials
There is no doubt that early groups would have been capable of making the necessary lunar observations. After all, Marshack’s work shows that hunter-gatherers were watching the skies. It is also unlikely that mobile hunting bands would have been blind to or simply uninterested in the opportunities that forward planning would bring. Mesolithic hunter-gatherers, operating in Britain between c.10,000-4,000 BC, would have benefited enormously from being able to anticipate seasonal change with accuracy.

The Warren Field pit alignment was discovered from the air by RCAHMS during the dry summer of 1976. Photo: Royal Commission on the Ancient and Historical Monuments of Scotland

The Warren Field pit alignment was discovered from the air by RCAHMS during the dry summer of 1976. Photo: Royal Commission on the Ancient and Historical Monuments of Scotland

Evidence for seasonal congregations of Mesolithic groups — at sites such as Vespasian’s Camp near Stonehenge (CA 271) and Star Carr (CA 282) — is increasingly being detected by archaeologists. These get-togethers appear to have involved far greater numbers of people than previously suspected. Whatever drew these groups together, be it ritual, an excuse for a party, or early networking opportunities based around marriages and alliances, timing would have been everything. It was essential that such gatherings occurred when the local area’s natural resources were sufficiently bountiful to support an influx of extra mouths to feed.

While the phases of the moon provide a convenient way of creating 29- or 30-day blocks of time, transforming this knowledge into a means of predicting the changing seasons is not straightforward. The earth’s orbit around the sun, which governs the seasons, does not coincide with a set number of lunar cycles. Instead the moon tends to undergo between 12 and 13 cycles in any given solar year.

This means that any attempt to keep track of the seasons using the moon alone will gradually drift ever further from true. An observer needs to know when to add or subtract an extra month to make good the time, or simply hit the reset button and start counting again. While it may sound complicated, once a group has acquired this knowledge it is relatively simple to construct a monument with a built-in reset switch. Vince Gaffney and his team believe that they have found possible evidence for just such a structure at Warren Field in Aberdeenshire.

First detected from the air during the infamously hot, dry summer of 1976, the feature consists of 12 pits set in a gently curving arc on sloping ground in the valley of the River Dee. The alignment created by the pits is around 50m long.


The pit alignment under excavation by the Scottish National Trust. Photo: Moira Greig, Aberdeenshire Council Archaeology Service

Evolving over several centuries, the pit alignment achieved its fully developed form in around 7,800 BC, pre-dating Britain’s first farmers by almost 4,000 years. This proof that Britain’s hunter-gatherers were constructing permanent monuments would be remarkable in itself.

Despite this modest scale, monuments sufficiently substantial to scar the earth are typically seen as the handiwork of settled farmers rather than roaming bands of hunter-gatherers, compelled to keep moving as natural resources were exhausted. At Warren Field, this prejudice appeared to be borne out by traces of a gigantic Neolithic hall lying immediately adjacent to the pits. Yet, when excavation of eight of the pits by the Scottish National Trust between 2004 and 2006 yielded enough material to establish a sequence of Bayesian radiocarbon dates, it produced a surprise: the pit group’s origins lay firmly in the Early Mesolithic period.

Vince Gaffney’s team have demonstrated how the combination of this pit configuration and their landscape setting could have accurately tracked the seasons, creating what is in essence a primitive calendar that the team have dubbed a ‘time reckoner’. If they are correct, it is the earliest such device known anywhere in the world.

Joining the dots
It was not just the date of the pit group that caught the team’s eye. The 12 pits were carefully cut so that the smallest two lie at either end of the alignment. From there the holes become larger until they reach pit 5, at the centre of the group. This cavity stands out by virtue of its size — a massive 2m across and 1.5m deep — and by having a conspicuously misaligned neighbour, pit 6, which breaks the gently curving arc observed by the other holes.

Pit 5 and pit 6 are also the only two in the sequence that preserved traces showing posts once stood within them. There is no sign that the other Warren Field pits ever held any kind of uprights. Instead, their fill contained large quantities of burnt material and — most intriguingly— stones that had been brought to the site from some distance away. It is a clear sign that special material was being deliberately placed in the pits.

Many moons ago
While the 12 pits could be taken to represent 12 months in the year, anyone attempting to use the monument as a forward planner still faced the problem of the inconsistencies between lunar months and the solar year. This cannot be corrected by a group of pits and posts alone, but the archaeologists noticed that Warren Field’s position within the landscape provided a natural reset mechanism.


This image shows the pits as excavated (bottom), with green indicating where they were later recut. Above that you can see the symbolic arrangement of the pits in relation to the Slug Road pass, as rendered on Google Earth. The sun indicates the location of sunrise on the midwinter solstice in 8001 BC, while the image of the waxing and waning moon shows how the pits could be subdivided to show the same sequence. Courtesy of the University of Birmingham

‘We started playing with the modelling of the monument and realised that a central, perpendicular alignment would strike the Slug Road pass. This is an important and conspicuous pass through the hills to the south-east of the site,’ Vince says. ‘Anyone watching from the Warren Field monument in 8,000 BC would have seen the midwinter solstice sunrise occur in the prominent notch on the horizon created by the Slug Road pass.’ The sun would have risen low down the eastern slope of the pass and then rolled up along its western slope before emerging from the gap. By creating a fixed point where an annual solar event can be witnessed, it becomes possible to prevent a lunar calendar from being cast adrift of the seasons.

‘Once a year, you effectively reset your clock by this observation of the rising sun,’ Vince explains. ‘Then you can start counting off the lunar months until the next midwinter solstice, when you start counting again. This means you are essentially kept on track for the entire year. It is not a complex thing. Once you have mastered the concept of lunar months that can be corrected on an annual basis, you essentially only have to track the months in order to make a functional calendar for that year. We can’t prove it, but we feel that the slight misalignment of pit 6 could be explained by this being the point of observation towards the midwinter solstice.’

This is an extract, but you can read the whole feature in CA 283.


  1. Helen Ball
    September 7, 2013 @ 7:44 pm

    This is such a fascinating article and theory.


  2. Douglas Scott
    November 30, 2013 @ 7:48 pm

    As someone with over 30 years experience of surveying alignments of standing stones, I don’t see how they can come to these conclusions as all they have is a curving line of pits. There is nothing from the pits that indicates any interest in the sun or moon at these times.


  3. David M
    December 9, 2013 @ 2:12 pm

    Has anyone done the maths to work out for how many days the alignment would have been visible – after all, the sun does “stand still” during the solstice. Also, what are the odds, given the weather, of actually being able to see it at all?
    Do hunter gatherers actually need this sort of thing? Are they not, given the vagaries of our climate, more likely to read signs in nature to determine when various foods will be available?


    • Another David M
      August 24, 2014 @ 7:09 am

      I agree. I find the statement “Mesolithic hunter-gatherers […] would have benefited enormously from being able to anticipate seasonal change with accuracy” quite risible. Why would they need a “calendar” when there are a myriad of natural signs of the changing of the seasons? The same claims are made for the Neolithic farmers – that they would need to know when to plant their seeds.
      The implication is that ancient man was incapable of noticing what nature was up to and needed some shaman/priest/lord to tell them. My advice to some archaeologists is to come down from their ivory towers and actually do a bit of hunter-gathering and farming.


Leave a Reply