ANNUAL REPORT 2006
D.J. Woolliscroft and B.Hoffmann
2006 has been the Roman Gask Project's tenth anniversary year and so we might perhaps spare a moment for the merest toot on our own trumpet: or at least to reflect on our achievements to date. The start of the Project was triggered by the realisation that the Roman frontier in Germany began later than had been thought, leaving the Gask to rocket to international importance as Rome's prototype fortified land frontier, and so the ancestor and prototype for the thousands of miles of similar systems that eventually grew up all round Rome's vast domains. When work first began, we thought that we needed only to refine our knowledge of an already reasonably well understood Roman frontier, and so gain a clearer understanding of the degree of evolution involved in the design of later systems. In the event, however, there proved to be vastly more to do than we had anticipated, for the simple reason that we find ourselves with such a rich archaeological resource (especially in Perthshire), which is capable of addressing deeper and more complex questions. As a result, our studies have become much broader than we had originally expected.
There is still much to do, but we have already cast a good deal more light on the anatomy, workings, rationale and dating of the Roman frontier. Our most important result has been plentiful evidence for a longer than expected life span for the occupation and this has raised the controversial, but ever more likely, possibility that the Romans were already ensconced in the area before the arrival of the famous governor Agricola, who had always been thought to be the first Roman general to reach northern Scotland. We have also made a significant contribution to studies of the interactions between the indigenous population and the incoming Roman army, with signs that relations may have been a good deal more peaceable than expected. In the process, we have run 28 excavations on 19 sites. We have conducted geophysical work on a further 13, including eight huge surveys, seven which took in whole Roman forts and their surroundings, whilst the eighth disproved what had looked to be another fort from the air. We have built up and catalogued an archive of almost 20,000 photographs, of which over 10,000 were taken from the air, and the aerial work has discovered literally hundreds of new sites, ranging from the Neolithic to WWII training trenches. At the same time, we have worked hard to ensure that our results enter the public domain as fully and as rapidly as possible. We have written a booklet and five full length books (and edited another). We have published some 38 academic papers, along with a number of archive and Internet reports, and have plenty of other material awaiting publication or in preparation, including a further three books. We have also tracked down and published the records of four past excavations, whose instigators failed to do the work themselves for whatever reason, and we have a further two such reports in preparation. We have created and maintained an award winning web site, which makes our research available, sometimes years before it appears on paper. We have provided training for over a hundred would be archaeologists from all over the world: mostly students, but including many local volunteers. We have given scores of public lectures and conference papers to explain our results, both in the Gask area, the wider UK and, as befits the Gask frontier's newly won international importance, in overseas countries, including the USA, Jordan, Spain, Ireland, Germany, Hungary and the Netherlands. We have also given numerous newspaper, radio and TV interviews and made television programs with 'Time Team' and 'Time Fliers'. In all, it has been an interesting start. Let us then pass on to the events of the current year.
Fieldwork
2006 has seen another busy field season, with another in our series of large, whole Roman fort geophysical surveys, this time accompanied by an excavation; a second geophysical survey on a possible new Gask frontier tower, and a continuation of our air photographic program, with both summer and winter flying. As always, our work has attracted volunteers from a wide geographical area, including the Gask region itself.
Bochastle
The Roman fort of Bochastle (NN 615078) is part of the so called "glenblocker" line: a series of forts between the Clyde and the Tay, most of which are set in or near the mouths of the main Highland glens. It was partly excavated by W. Anderson and the Glasgow Archaeology Society between 1949 and 1953 and as a report was produced with commendable promptness (TGAS 1956), the site should have been well understood. Sadly, however, the published account is (to say the least) not a model of clarity, made worse by poor illustration. The work was sufficient to prove a later 1st-century, Flavian date, but it also claimed a number of suspiciously unusual features. For example, the excavator claimed that the fort had only a single defensive ditch, whereas air photography reveals a far more normal two. He also asserted that the north rampart angled off towards the south, to the east of the north gate and that the fort lies markedly off centre within its ditch system, with the west rampart only 1.8m from its ditch, compared with 22.6m in the east. The ramparts appeared to turn inwards to provide deeply recessed gates on all four sides and, as few internal features were identified, the excavator came to believe that the adjacent River Leny had eroded the fort at some point, removing all trace of buildings. This final conclusion was particularly startling as the one reasonable piece of area excavation that was conducted in the interior, recorded a floor and a line of posts, which might have been part of a barrack. The report appears to give this discovery little significance, however, and did not even bother to provide a plan. Likewise other remains towards the north-east corner were brushed over with hardly a word and it is to be regretted that so important a site could be treated in such a cavalier manner at a time when archaeological technique was already well advanced.
In an attempt to clarify matters, the Gask Project conducted resistance and magnetic surveys covering an area of over 22 acres, which took in the whole fort and its immediate 
surroundings, along with the north-eastern quadrant of a large accompanying temporary camp. The work formed part of our longer term series of fort surveys, which eventually plans to cover all of the forts north of the Forth and Clyde, and it took us over the half way mark for the program: with seven forts now finished and a further six still to go.
The results provide a significant advance in our previous knowledge and firmly disprove many of the claims made in the excavation report. They confirm, for example, the aerial indications that the fort had two ditches on all four sides and show an upcast mound outside them. There is no sign that the northern defences made the course change the excavator claimed: instead the defences were simply eroded away by the river at the north-east corner. The survey also confirmed aerial indications that the ditches formed a variant on the normal Flavian "parrot beak" gate entrance breaks which has also been seen at the next "glenblocker" fort to the west: Malling. In a standard parrot beak, the outer ditch turns inwards to meet the inner on either side of the entrance, but at Malling and Bochastle, both ditches turn inwards before meeting, to create a pattern that more resembles an eagle's beak. The fort ramparts showed well as a series of high resistance bands and there was no indication that the east rampart lay, as claimed, at any greater distance from the ditches than the others. Instead there was a uniform separation of c 2.6m all round. One aspect of the excavation plan was, however, strikingly vindicated: the rampart re-entrants, which do indeed occur at all four gates and, although not completely uniform, these are quite as deep as claimed: ranging from 19m - 26m. Similar features have been found at a number of other 1st-century forts in the north, with Cardean, Drumquhassle and Strageath all showing gates of a similar design, albeit not on all four sides. It is thus possible that gates of this type might prove to be more common than is currently suspected, and may eventually be seen as a diagnostic late 1st-century feature. Tactically, they fit well with the parrot/eagle beak ditch breaks, to create what amount to funnels that could be guaranteed to cause confusion amongst even an ordered rush on the gates. Because of their design, an attacking force would have been able to pass through the outer ditch on quite a broad, c 25m wide, front. It would then find itself rapidly compressed as the outer ditch ran in towards an
inner ditch break only c 15m wide. Some of its outermost members might even be pushed into this ditch, but the others would be forced inwards, causing confusion in the ranks at a time when they were already under fire from the fort. They would then pass down the rampart re-entrant, becoming still more crushed and disordered, to a fort entrance just c 6m wide. There they would be faced with the barrier of the gate itself, whilst under enfilading fire from the rampart funnel, and with their escape route blocked by their own comrades forging on from behind. The result would be a killing ground with the funnel acting as a trap: a very neat, if unpleasant, piece of design.
There were firm indications that the interior is much better preserved than the excavator would have us believe. For the road between the east and west gates showed strongly as (more importantly) do signs of internal timber building foundations. In particular, two large courtyard structures from the main range were visible immediately south of the east-west road, with even room divisions being detected. Three of the fort's four corners have survived, although that in the north-west has only just been spared by the river and a only small part of its south-western counterpart projects from beneath a (now disused) Victorian railway embankment. Nevertheless, this is enough to provide a more accurate size for the site, which now measures 148m (e-w) x 178m (n-s) over the ramparts and 170m x 197m over the outer ditch. Making allowance for the gate re-entrants, this gives an internal area of c 2.2 ha (5.4 acres), which is slightly larger than had been estimated from air photographs, but still smaller than many other Flavian forts in the region.
The temporary camp lies to the west of the fort and was previously known only from the air, as the only past excavation, which was designed to check the line of the north ditch, failed to find it. No 
gates have been detected on the east and west (long axis) sides, but there are 'Stracathro' style gates in the north and south: a type usually ascribed to the Flavian period. The site is unusually large for the type, however, for although its exact dimensions required clarification, it appeared to measure c 530m (n-s) x 365m (e-w), an area of c. 19.4 ha (47.8 acres). The difficulty in determining the camp's size derived from an oddity on its west side, where air photographs have shown up to three widely spaced ditches. There had also been aerial indications of a small rectangular enclosure attached to the southern end of the east rampart which resembled an annexe. Camps with annexes are not unusual in Scotland (e.g. Craigarnhall a few miles to the SE), but these are usually still larger sites of c. 26 ha (64 acres) with titulate gates and, at present, no other Stracathro camp has produced evidence for such a structure.
The geophysical survey scanned the camp's north-east corner, along with 124m and 258m (respectively) of the north and east ditches, which all lay on exactly the lines predicted by air photography. The camp's east ditch intersects the south-west corner of the fort but, as it approaches, it becomes obscured by the fort's upcast mound and may well be overlain by it. If so, the camp would be earlier in the sequence. In the past it has been suggested that the camps outside a number of the glenblockers provided accommodation for the troops who built the forts, although they seem unduly large for the role, but this sequence might cast doubt on that theory as it seems improbable that the soldiers would have dug a fort ditch through their own, still occupied, camp.
Two trenches were excavated on the camp defences. In the west an attempt was made to determine which (or which combination) of the three possible ditch lines 
was the real camp ditch. In the event, all three proved to be Roman military style, V shaped ditches and two had been re-cut. No dating evidence was recovered to prove beyond doubt that all were indeed Roman, but it seems likely and, if so, they would suggest a complex occupation history, especially as temporary camps do not as a rule have fort-style, multiple ditch systems, so that it is unlikely that two or more of these cuts were ever in active use together. On the other hand, the second trench, which investigated the possible annexe ditch, found it to be a natural feature.
The work also cast light on a number of non Roman periods. The outermost camp ditch in the west had cut an earlier, flat bottomed ditch which did not appear likely to be Roman. It might, instead, be an outwork of the nearby Dunmore hillfort. Both trenches also produced plentiful evidence of post-Roman (probably 18th and early 19th century) metal working, with furnaces, iron blooms and furnace waste all appearing. A series of early modern lime kilns were surveyed, as was a large, stone revetted, oval platform, in the centre of the camp, which had been cut into by (and so predated) the buildings of Bochastle farm, and may be a bank barn or a Medieval motte and bailey, similar to a number found elsewhere in the same glen in Perthshire.
Woodhead
This site was (and to a degree remains), something of an enigma. It stands atop a marked ridge just south of, but well 
above, the River Tay, near Wolfhill (NO 143346), and has a superb field of view which takes in some 30 km of the Highland fringe. Air photography has revealed two distinct features on the site. The first is an unenclosed settlement of probable Iron Age date (fig 5, arrows B), but it was the second, a double ditched enclosure (fig 5, A), that attracted the Project's attention, particularly during the 2005 flying season when it made a better than average showing.
Since the 1980s, the Gask frontier has been known to run from the fortlet of Glenbank in the south, to the fort of Bertha on the Tay: a run of c. 37 km. But this has never seemed likely to
be the whole story. Glenbank seemed an unlikely spot for the system to stop and it is anyway more usual for Roman frontiers to end at forts rather than fortlets. As a result, it seemed all but certain that more still awaited discovery in the south. The northern end was more difficult to predict, since the Tay seemed a logical stop line and Bertha, as a fort, was a plausible terminus. Nevertheless, there are additional forts beyond the river, leading far up Strathmore and there is also one more tower, at Black Hill. This is probably close enough (1.5 km) to the fort of Cargill to be treated as a satellite but, if so, it would be unique in northern Scotland, thanks to growing doubts over what had been thought to be another satellite tower near Fendoch fort. It might thus seem all the less likely to have stood alone. As a result, the Gask Project has long been looking for evidence for additional sites at both ends of the established line. We have in the past detected (but then rejected) a number of candidate ring features close to the Roman road in the south and have another under investigation at present, but Woodhead is the first we have considered in the north.
From air photographs the site scales off at around 26m in external diameter and its inner ditch is noticeably subrectangular, whilst the outer is more circular. In other 
words, it bears a striking resemblance to the double ditched watchtowers at the southern end of the Gask line. It also sits close to the line of a putative Roman road from Bertha, which is mentioned by 18th century antiquarians and marked on two 19th century maps of the area. Moreover, its field of view might have been purposely chosen to allow signal communications via the visual techniques used by the Roman army. For it has the legionary fortress of Inchtuthil, the key site of the entire northern occupation, in full view and is also intervisible with the tower at Black Hill, which could link it to the next fort to the north: Cargill. Unfortunately, the Woodhead feature's identity could not be confirmed from aerial data alone, for the simple reason that it has never shown in its entirety. The site stands close to the northern edge of a field and the plough headland interferes with cropmark formation. As a result, no air photograph has revealed the northern side of the outer ditch and only one (CUCAP neg: DG63, of 1949) shows that of the inner ditch. The southern section, visible on numerous air photographs, does not show the usual entrance break of a Roman tower and, although the Cambridge picture does reveal a narrow, c. 1.8m, break in the inner ditch at its closest point to the modern field boundary, the position with regards to the outer ditch was unknown. As a result, although it remained possible that a full tower entrance lay in the missing northern section, there was no firm evidence either way. In an attempt to learn more, the Project conducted a geophysical survey of the site. However, although valuable information was gained, the results still retain something of the ambiguity of the aerial data.
The 2.3 acre survey revealed the entire circuit of both of the double feature's ditches, which allowed its size and shape to be determined with much greater accuracy. Interestingly, both ditches produced high (rather than the normal low) resistance readings, but this is quite a common phenomenon in the area and has appeared on a number of previous Gask Project sites, where excavation has confirmed the cropmark evidence that the anomalies did indeed mark ditches and not more usual high resistance features, such as wall foundations. One particularly marked case was the Gask tower of West Mains of Huntingtower and in all the examples met with to date, the explanation was that the ditches contained a high stone content, thanks to being used as stone dumps in the past. In other words they were backfilled by being turned into what amount to inverted clearance cairns.
The outer ditch was a slightly subrectangular oval in shape: measuring 26.5m in external diameter from SW-NE, but 28m from NW-SE. Its resistance plot is markedly more pronounced than that of the inner ditch, which suggests that it is both wider, at c. 2m (as opposed to c. 1m), and probably deeper. The inner ditch was, as expected, significantly more subrectangular in plan, but it was also more uniform in diameter, at 15.5m all round. The two were not exactly concentric, however, for the centre point of the inner ditch lay about 1m to the NE of that for the outer. In the NNW, at the closest point to the modern field boundary, both ditches seem to show a narrow break, of around 1.5 - 2m for the inner ditch and 2.5m for the outer. The former agrees well with the 1.8m break scaled off the 1949 Cambridge air photograph. There was, however, a degree of interference from the field headland at this point, so that it was not possible to be absolutely certain that the breaks represent genuine archaeological features, and a higher resolution survey conducted the following day proved unhelpful thanks to waterlogging caused by heavy overnight rain. As a result, the double ditched enclosure's identity remains frustratingly uncertain. If it had been firmly shown to have entrance breaks in its ditches it would have seemed almost certain to be a Roman tower because, as mentioned, the rest of its morphology makes it a virtual double of the southernmost four towers on the Gask frontier. All four of these sites have now been closely studied and all are known to have an almost identical configuration to Woodhead: with fairly circular outer ditches and subrectangular inner ones. Their dimensions also match Woodhead almost exactly.
|
|
||
| Table 1 | ||
| Site | Outer Ditch diameter |
Inner ditch diameter |
| Greenloaning Gask tower | 24.70m | 15.54m |
| Blackhill Wood Gask tower | 24.83m | 16.47m |
| Shielhill South Gask tower | 25.55m | 15.27m |
| Shielhill North Gask tower | 24.00m | 15.72m |
| Woodhead | 26.50m | 15.50m |
|
|
||
Table 1 shows the comparison, and it can be seen that Woodhead's inner ditch falls perfectly into the range shown by the Gask towers, whilst its outer ditch is only outside that range by a matter of centimetres. There are differences: for example the southern Gask tower inner ditches were more substantial than their outer ditches, whereas the situation at Woodhead is reversed. Likewise, the possible entrance at Woodhead, if real, would be slightly narrower than the 3 - 3.5m entrances found on the Gask. But these are largely matters of detail and certainly not enough to rule Woodhead out as a tower. Of course it could be pointed out that the southern towers are a long way away (37.5 - 42km) and that the closer, more northerly, Gask towers are all single ditched. Consequently, it might seem unlikely that a double ditched tower would be build so far north. But there is evidence that the Gask was constructed as a series of individual building sectors in which installations with slightly different design traits were built by different military units (probably legions). As on other Roman frontiers (e.g. the Antonine Wall) it is thus perfectly possible that the same unit may have built more than one sector: be it contemporaneously, or as part of a sequence, so that its particular building idiosyncrasies might be seen on widely separated parts of the same system. Woodhead's field of view would also make it a perfect tower site, because of the possible signalling links already described and its superb potential as an observation post. Nevertheless, the lack of certain proof for an entrance break in both ditches means that doubts must still remain and there are other site types that this could be. One possibility is a Pictish barrow, and only excavation can answer the question with certainty.
Hopefully we will be able to conduct such work in the not too distant future, because the historical knowledge to be gained could have connotations well beyond the site itself. At present, the pre and post-Roman Iron Age in this area is inadequately understood, and the prospect of studying both an unenclosed settlement and a possible Pictish monument together is highly attractive, especially if one proved to evolve from the other. On the other hand, the potential ramifications of a confirmed Roman tower could be still greater, for it would radically change our understanding of the whole Gask system. Woodhead lies far enough from a major garrison fort that such a small, lightly manned installation would be highly vulnerable. Worse still, it could not see (and so directly signal to) the nearest major site: Cargill (which is closer and far more accessible than Inchtuthil) and, although, as stated, it could be linked to the fort via Black Hill (at a range of 5.5 km), this might be thought a somewhat tenuous connection on which to stake soldiers' lives. If the site was a tower, therefore, it seems unlikely to have stood alone and its confirmation would thus be of major significance, because it would strongly imply that we should also be searching for others. Moreover, the possibility of a Roman road, as claimed by earlier workers, might even argue for a continuation of the full Gask frontier system as far as Cargill, if not further, and the prospect of discovering 13 or more kilometres of new Roman frontier would be a major discovery which would dramatically alter our picture of late 1st-century Scotland.
The survey also covered part of the unenclosed settlement and the data, when combined with rectified air photographs, has allowed us to produce the first detailed plan of the site. Parts of at least ten ring features were detected, most of which are likely to represent roundhouses. Some overlap, however, and thus cannot be contemporary. This is perfectly normal, and excavations on such sites often show that they actually represent a small number of structures which have been repeatedly rebuilt in slightly different positions, rather than a single contemporary concentration. The Woodhead settlement may thus have been smaller than it appears, at any given time, but also, perhaps, rather longer lived. The relationship between the unenclosed settlement and the possible tower remains ambiguous. The latter's outer ditch certainly overlaps one of the ring features and, although it is always dangerous to draw stratigraphic conclusions from remote sensing data alone, the fact that the ring could not be traced across its interior might suggest that the double ditched enclosure came later. Whatever the case, however, the evidence that the unenclosed settlement had at least a reasonably long life, means that a proven lack of contemporaneity between the possible tower and any one of its component structures does not rule out the possibility that it could still have coexisted with others and, again, only excavation could settle the matter beyond doubt.
Air Photography
2006 was a curious, but exciting flying season. An extremely dry summer produced what should have been ideal cropmark conditions, and in many areas this did indeed prove to be the case. Certain sites showed in spectacular form: better than ever in the history of aerial observation. Our old friend, the fort of Cardean, for example, produced an almost perfect plan 
with clear signs of structures all over its interior. Better still, further external Iron Age features were revealed to add to the sites detected by earlier aerial work and our own geophysical survey. For example, a 19th century report describes a souterrain to the east of the fort, although its exact location remained uncertain. But the feature was clearly visible this summer and produced a considerable surprise, because our pictures show what appear to be two additional souterrains close by. 
The fortress of Carpow, on the Tay/Earn confluence, showed almost as well, with the western defences making a rare appearance, and lines of internal postholes marking out the previously little understood barrack lines. Elsewhere, large numbers of new native sites came to light for the first time, or showed previously unknown details, and a particular concentration was discovered around Coupar Angus, which extended the size of a previously known unenclosed settlement to the point where it could almost qualify as a town. In other areas, however, we were given an excellent demonstration of how poorly the exact mechanisms of cropmark formation are currently understood. For, despite what appeared to be equally ideal circumstances, almost nothing was visible even on sites that normally show in poor cropmark years. Nevertheless, the harvest was still exceptional. In all, 105 new sites were discovered, mostly as cropmarks, but with others appearing as shadow marks during a single winter flight in January. Over 1,900 photographs were taken and most had been fully catalogued, digitised and mapped by the time of writing.
As in previous years, our flights were made from Scone airfield, and we are, as always, immensely grateful to Bill Fuller for volunteering his services as pilot. Bill has also continued to work on creating a searchable computer database of all of our air photos to date, which will form a far more flexible means of access for outsiders to use than our present catalogue listing. He has now nearly caught us up: no mean feat with an archive of over 10,000 pictures. The considerable investment in new cameras made in 2004/5 has also paid off, with the technical quality of our images being noticeably improved over those produced by the already high quality equipment used previously.
Collaborations
The Gask Project has again been able to work with a number of other scholars to gain additional information from our own sites and those of others. Dr's J. Huntley and S. Ramsay have continued to handle our environmental evidence. Prof E. Mackie has analysed our querns and we have worked closely with geophysicists David Hodgson, Susie Moore and Peter Morris. We have also worked with Gavin Lindsay, a final year student from the University of Durham, who has been producing computerised viewshed analyses of the fields of view of the Roman sites in the Gask area which have shone useful additional light on the rationale behind the layout of the system. As in previous years, we have involved interested local people in our field work and quite a number of local volunteers helped with our excavation and survey at Bochastle. We are also involving volunteers from a number of local historical societies in a new field and archive survey of the area around the legionary fortress of Inchtuthil (see below).
We have also begun two international collaborations, firstly with Perth's twin city, the Roman fort town of Aschaffenburg, on the German frontier and, secondly, with archaeologists working on the Dutch and German legionary fortresses of Nijmegan and Bonn. More details are provided below.
Publications, Outreach and Publicity
As ever, the Directors have continued to give lectures to a variety of academic, student and amateur bodies: 20 in all this year and, as always, we have made particular efforts to respond positively to any requests to speak in Perthshire, with seven talks given there this year. The latter seems only fair given the long term sponsorship we have received from the county, even though local societies cannot usually cover all of our travel expenses. To get round this, we usually try to group the talks so that the costs are shared, or to schedule them to coincide with field or archive work in the area. Elsewhere, both directors gave papers to the annual Roman Army School at the University of Durham and to the triennial, International Congress of Roman Frontier Studies, which was held in the Spanish legionary fortress city of Léon in September. David gave a paper to the Hadrian's Wall conference in South Shields, whilst Birgitta spoke in Aschaffenburg, in Germany and we have also spoken widely in other parts of Britain. Media activity has also continued with Birgitta making another appearance on Time Team and David, a number of radio and newspaper interviews.
As always, 2006 has seen a number of Gask Project publications, of which the most important were two new 
books. The first was a full length account of the archaeological evidence for the 1st century Roman occupation of Scotland north of the Forth and Clyde. The last time this was done was in O.G.S. Crawford's much loved (if less than catchily titled) "Topography of Roman Scotland North of the Antonine Wall", published in 1949. But a great deal of work has been done since, not least our own fieldwork and the explosion in aerial activity, so the book is now hopelessly out of date. Our own book: "Rome's First Frontier" (Tempus Books), takes a broadly similar approach to Crawford. It tries to allow (indeed encourage) readers to form their own concept of events by keeping fact and interpretation as firmly apart as possible. The first half provides a detailed, up to date and lavishly illustrated summary of all the currently available archaeological data but, whilst the second offers our own views on what this means, we hope that readers will use the information provided in the first section to argue with us and develop their own (hopefully better) interpretations. So far the book seems to have been well received, with good sales and a number of glowing reviews: one called it a "masterpiece" and "epoch making". Although one Agricola die-hard amused us by dismissing installations being rebuild from the ground up as "subtle readjustments". Our sponsors the Perth & Kinross Heritage Trust very kindly set it off to a good start by organising a launch party in Perth Library, accompanied by a lecture on Tacitus' place as a writer and orator by Birgitta.
The year also showed up one of the many advantages of the Project having a German speaking co-director. For our second book, "Die Römer in Schottland", is in German and stems from a collaboration we have begun with Aschaffenburg, as part of the city's twin town arrangement with Perth. As both lie on Roman frontiers, the scope for archaeological co-operation is obvious, and the book was written to be published in
association with an exhibition on our work in Aschaffenburg Museum this April, again launched with a lecture by Birgitta. The book is not simply a German translation of the Tempus volume; it is a wholly different work which covers a wider geographical area and a longer time scale to give a broader brush account of the wider Roman occupation of Scotland as well as just the Gask. The book was very nicely produced in full colour, although, thanks to an infuriating publisher's error, David's name was omitted from the by-line and the book was assigned to Birgitta only.
The year has also seen the publication of a number of academic papers. David has published two on Roman/native relations: one in the proceedings of the last Roman Frontiers Congress, which was held in Hungary in 2003, and a second in the Hadrianic Bulletin. Birgitta published her work on the effectiveness of Roman clay sling shots (as found during our work at Drumquhassle fort) in the Arbeia Journal; a study of Roman melon beads (a common find in the Gask area) in a book in honour of her former Ph.D supervisor; a study of the origins of Roman artefacts from native sites in the Gask area, in the Hadrianic Bulletin and a glass report for the Iron Age site of Aves Ditch. On a more popular level, we also published an account of our East Coldoch native site excavations in "Current Archaeology".
A number of other papers have been readied for publication during the year. A report on our Woodhead survey has been completed for the Hadrianic Bulletin and the Bochastle report now only awaits its specialist (mostly environmental) reports. The latter will eventually appear alongside our other whole fort surveys in a book which we have agreed to produce for the British Archaeological Reports series. Our Frontiers Congress papers will both appear in due course in the published proceedings, as will David's paper to the Hadrian's Wall conference, and Birgitta has submitted no less than five new glass reports. She has also made a wholly new departure in taking over the editorship of the "Hadrianic Bulletin" which, despite its name, is a journal of Roman military studies, with an international distribution. She has already presided over her first issue, which went out just before this was written and is currently involved in an energetic overhaul which is turning this formerly fairly sleepy periodical into a more thrusting, ISSN numbered and peer reviewed journal, hopefully without loosing its long tradition of accessibility to the non specialist. As ever, our web site continues to be kept up to date and our Woodhead and interim Bochastle reports should go on line at the same time as this report.
When considering outreach it is easy to think just of training and fieldwork opportunities given, site tours and talks to local groups, but it is also worth mentioning the considerable time we spend in correspondence with local historians, students, school children (and their teachers) and just normal interested citizens who contact us for (or with) information, advice on their own projects, or when worried that building and other activities might impinge on historical remains. This year has been particularly busy in this respect, including a positive deluge of messages from people worried about the prospect of quarrying near the famous Roman fort and temporary camps at Ardoch.
Sponsorship and Acknowledgements
The Project continued to be sponsored by the Perth & Kinross Heritage Trust, whose support has, as always, been indispensable and is very much appreciated. In 2006 the Trust funded our air photographic flying program and the purchase of third party air photographs, along with the accommodation for the Woodhead survey, environmental work and a set of geological maps of the Gask area. 2006 was the final year of our current funding tranche from the Trust, but we have applied for a renewal and are living in hope of its being granted.
In addition to this long term funding, we have continued to attract one off grants for specific programs and this year's survey and excavation at Bochastle was funded by a generous award from the Roman Society. We are grateful to our long standing corporate sponsor (which continues to insist on anonymity), which has again provided material support and we have also received a number of smaller donations from speaking engagements, voluntary bodies and private individuals and, as usual, the royalties from our new books will go to the Project.
Finally, the Project continues to owe thanks to the farmers and land owners who have allowed us access to sites, to Dr David Simpson, who again provided medical services during our fieldwork, and to Peter Green who has continued to do a wonderful job of maintaining our web site. Tayflite Ltd let us play with their aeroplanes. Mrs Hillary Fuller spoilt us rotten whenever we flew. Andrew and Eleanor Graham once more provided our dig accommodation and we continue to receive much valuable help from the staff of the J.K. Bell Library whenever we use the Perth archives. Finally, as always, we are grateful to our many field volunteers, especially our long-standing trench supervisor, Keith Miller, and his more recently acquired colleagues, Paul Murdoch, Lindsay Farquharson, Mark Sephton and geophysics supervisor Rachel Hunt.
The Future
So far, the Project has been organised in two five year phases, each of which has seen a distinct change in our research priorities. This does not mean that we abandon previous interests, but it does keep things fresh and it ensures that we are constantly covering new ground and getting useful new angles, when it is all too easy for longer term research projects to become bogged down in minutiae. For our first five years, we concentrated almost exclusively on the Gask fortifications and their history. The last few years have seen a much greater emphasis on forts and on the native side of the equation and work on all of these areas will continue, but it is now time for another shift and we want our third phase to give more attention to a site that we have all but ignored up to now: the vast, 50 acre, legionary fortress of Inchtuthil. This is, in many ways, the key site for the entire occupation, being the base of the only legion in the Gask area. These wholly Roman citizen, heavy infantry units were the elite of the Roman army and, with an official strength of 5,500 men, this one unit accounted for around a third of the entire northern garrison. Indeed, its vital role is amply demonstrated by the fact that the ultimate abandonment of the Scottish conquests in the mid 80s AD is thought to have been triggered by the withdrawal of a legion from Britain following a series of Roman defeats on the Danube.
The fortress itself, and the group of temporary camps and smaller fortifications around it, are at least reasonably well understood, thanks to aerial work and two excavation series, by Abercrombie and then Richmond in the 20th century. We are thus not planning excavations of our own, at least unless some specific need arises, although we have combined the two existing excavation plans, something which oddly had never been done before. But that still leaves plenty to do. Firstly, the area immediately outside the defences has hardly been touched by excavation and does not respond well to aerial photography. For example Abercrombie found two large rectangular stone buildings just east of the fortress and 18th century antiquarian accounts report the discovery of a bath building at some unknown point to the west of the site, none of which have ever shown from the air. Consequently, there may well have been other external structures serving industrial and recreational functions, or even a civilian settlement. The fortress itself is probably too large and too well known already to be worth a complete geophysical survey, but an external survey would thus appear to be very well worthwhile.
In many ways, however, the surrounding area might be a more rewarding study target than the fortress itself, for the coming of the legion will inevitably have had a considerable local impact. We also need to know more about its ancient setting. The site lies on a promontory just north of the Tay, but there is plentiful evidence that the river has shifted course many times and that it sometimes puts the fortress to its south. There has obviously been at least one such change since the site was occupied, because the northern defences have been eroded away by the river, which has also left a series of oxbows around 
the promontory's west and north sides, It is thus perfectly possible that the site was actually south of the Tay in Roman times and, even today, it can become an island when the river floods badly. Indeed it is reported to be so as we write. Obviously, it is difficult to gain a detailed understanding of the fortress' intended role without knowing where it was in relation to such a large river and the issue will also have relevance for the site's communications. At present, strange as it may seem, no Roman road links have been traced to Inchtuthil, except for a relatively short length of road which runs north towards the quarries which produced the stone for its ramparts. It seems unlikely that such a vital unit would really have been left cut off by road, and we have been searching for possible lines from the air for some time. Without knowing which side of the river the site was originally built on, however, it is obviously much harder to know where to look.
More importantly, legions were usually assigned a territory (variously called a saltus, territorium or prata legionis,) around their base, whose agricultural, mineral and other resources were placed at the army's disposal. This much is well understood, but we still know very little about how large these territories were; whether they were unified zones, or discontinuous blocks with specific essential resources, not to mention how they were organised and what impact they had on the local inhabitants. We also have little idea of how soon after a legion's arrival they were established or how quickly their exploitation developed, and this is where Inchtuthil might have particular value, because it combines a short life span, with a largely non built-up modern environment, which should allow us to study the early stages of a legionary territory's development over a wide, clearly visible area. As a result, we have already begun a study of the whole Inchtuthil region. At present this has involved the start of a particularly intensive aerial survey of a zone within a 10 mile radius of the fortress, to look for differences in the population of native sites, and we are also studying the area's flooding and land use patterns, along with the availability of mineral deposits (especially metals), building stone and potting clays. This work, especially the aerial side, will continue, but future seasons should also allow us to move on to field studies, such as fieldwalking and excavation and, of course, our possible tower at Woodhead is very much part of this study area. Finally, Inchtuthil cannot be looked at in isolation as it was just one of many legionary fortresses set around the Empire as a whole. As a result, the survey will have a firm international dimension, as we have already agreed to co-operate and exchange information with archaeologists planning similar work around the fortresses of Nijmegan in the Netherlands and Bonn in Germany, and we hope to involve others in the future.
In addition, 2007 should see another in our series of whole fort geophysical surveys and, at present, we have our eye on Bertha, just upstream of Perth on the Tay. We are also hoping to excavate on that site, if consent can be obtained, and to run another dig to determine the relationship between the Gask frontier road and the huge Innerpeffray West temporary camp. Our air photographic program will continue and we also hope to conduct geophysical work on a ring feature discovered at the southern end of the Gask during our 2006 flying season, which might be another tower candidate.
Out of the field, we plan to finish the Bochastle report and to prepare the final reports for our excavations on the tower at Garnhall, Glenbank fortlet and the East Coldoch native settlement. The Directors will continue to give public lectures where invited. Four have already been booked, along with further talks to the Roman Army School, and others will no doubt be scheduled as the year progresses. We also plan to continue the association with Perth's twin town, Aschaffenburg, as comparative work between the Gask and the German frontier and the opportunity for the Project to compare notes with German scholars has real and obvious research potential beyond the diplomatic and public relations side of a town twinning arrangement. It might also attract tourists to the Gask. For 2007, Aschaffenburg have agreed to fund a field study of the likely signalling arrangements on the frontier around the town, which will make a useful comparison to our results on the Gask. This is to be followed by an experimental demonstration of the signalling techniques and equipment used by the Roman army, which will be covered by German television.
D.J. Woolliscroft and B.Hoffmann
Directors: The Roman Gask Project
SACE
University of Liverpool.
www.theromangaskproject.org.uk
