The site of Fiskerton in Lincolnshire, in the fenlands of eastern England, is one of a number of wooden trackways or causeways built during the Late Bronze Age and Iron Age (late second and first millennium BC) across marshes, shallow lakes or rivers at various locations in Britain and continental Europe. Whether we might wish to classify them as “bridges” (connecting dry land) or “jetties” (for example, leading out onto platforms over open water) is not always clear from the available evidence, but what they have in common may have to do less with this (to us) pragmatic aspect of their function as with their strong association with votive deposits connected with water. Underneath and along both sides of the Fiskerton causeway, which was supported by two rows of upright posts, were scattered literally hundreds of what appear to be votive offerings, including precious weapons such as daggers and swords as well as more mundane items of metalwork such as hammers.
This is interesting enough in itself, but what sets Fiskerton apart is evidence that the causeway underwent major refurbishment at regular intervals. One of the advantages of ancient timber artifacts and constructions, where they remain to be examined by the modern archaeologist, is that they are susceptible to dendrochronology (tree-ring dating). This is a technique that can specify the date of felling to a particular year, and even season. Thus we know that the Fiskerton causeway was periodically maintained by adding new posts, and that the felling dates span a period of over a century, from 457 b.c.e. to 339 b.c.e. It appears that the felling took place predominantly in the winter or early spring, and that few if any timbers were seasoned before use—in other words, they were put into use within a year. Remarkably, the causeway uprights seem to have been felled—and, by implication, placed—at regular intervals of between sixteen and eighteen years.
Given the evident sacred significance of the site, an obvious possibility is that the regular episodes of major refurbishment—likely accompanied by due ritualistic observances—were astronomically regulated. Archaeologist Andrew Chamberlain has examined the possibilities in detail. A prime candidate would seem to be the Saros cycle. This represents a period after which the sun, moon, and earth return to almost exactly the same configuration in space. The upshot is that, at any time, the moon’s phase and position in the sky are more or less identical to what they were eighteen years and eleven days before. In particular, if a lunar eclipse occurs once, then another is likely to occur eighteen years and eleven days later. Chamberlain concluded that the major episodes of construction were indeed correlated with total lunar eclipses—those associated with two particular Saros series (nos. 46 and 47).
The implication here is that the people who used Fiskerton could predict eclipses as opposed to merely reacting to them. Unfortunately, this appealing conclusion may not be as clear-cut as it seems. For one thing, the fact that one lunar eclipse may be prominently visible does not necessarily mean that the next one in the given Saros series will be prominent as well; the successor may, for example, occur below the horizon during the daytime. Second, many different Saros cycles, running concurrently, produce eclipses from one year to the next. Third, some lunar eclipses will be missed on cloudy nights. In the absence of any form of counting or recording, the occurrence of total lunar eclipses observed from any given location could not seem anything but entirely irregular. There are some short-term cycles that, once recognized, will yield partial predictive success, but only after very careful recording maintained systematically over several generations (as among the ancient Babylonians) can longer-term cycles such as the Saros become evident. And even then, the Saros is not the only cycle that might be recognized: the ancient Chinese, for example, identified a shorter one of 10 years and 334 days, while the Maya discovered a longer one of 32 years and 272 days. While we should not underestimate the capacity of people in Iron Age Britain to keep records had they been so motivated, we have no direct evidence of this.
It remains possible that some of the bursts of construction at Fisker-ton were carried out in response to lunar eclipses, and indeed there is some evidence from other European trackway sites of correlations with various lunar eclipses that might support this conclusion. However, in this case the apparent regularity of the interval between the bursts of activity at Fiskerton would have to be fortuitous, since people would have had no way to distinguish only those eclipses from particular Saros series. Further evidence and statistical analysis may clarify some of these issues.
Despite all these imponderables, Fiskerton has a wider significance. One of the most important things about the sky as a cultural resource is that the regular cycles of the celestial bodies provide a way keeping the timing of various activities in tune with nature. Yet archaeoastronomers are seldom faced with evidence that bears directly on this, since the time resolution of most archaeological data is so imprecise. (Some have tried to use the astronomy to do the dating, but this is dangerous because it can so easily result in circular argument.) Even the best radiocarbon dating can seldom tie down a particular action to much better than the nearest century, and often the uncertainties are much greater than this. Fiskerton represents a rare example of a prehistoric site—certainly the first in Britain—where the timing of several successive bursts of activity can be tied down to the very year, if not the season. This enables us to ask not only whether they were correlated with the regular cycles of the heavens, but also whether they may even have had some connection to actual celestial events such as particular eclipses. This sets an important precedent for other places where we are lucky enough to be able to recover timber artifacts and monuments.
