The project reported here was devised with the ambition of locating the site of the battle of Fulford. 

The diverse work outlined in this report, plus the subsequent research, analysis and interpretation, have revealed an area which is the probable battle site of 1066 and now merits much more investigation. We now think we know what we are looking for, and where it is located, and this will permit the methods to be improved and the work to be focused. Our work, plus all the analysis, has allowed a confident hypothesis to be set forward.

Another project is now required, not only to test the work that has been done but to undertake a number of additional investigations which will allow a more precise interpretation about the location and course of the battle.

The project has worked under many quite unnecessary restrictions imposed by those hoping to build housing on the nearby land and, much more critically, to construct the access road directly along Germany Beck. 

Much of out intended investigative work was blocked. It was especially frustrating when follow-up work on exciting finds was prevented once the information about what we had discovered was disclosed. Precisely how such a cynical embargo could be placed on our work, and the supine way that the planning system reacted to this, is discussed elsewhere.

It is a pity that there are so many loose ends. These have forced me to speculate instead of investigate becuase access was denied to the land. If access had been allowed, I believe that the location of the battle could have been put beyond any reasonable doubt which, I suspect, is precisely why we were not allowed to pursue the work.

These omissions must now be repaired. This work must be undertaken before the site is disturbed or destroyed, as every piece of planning guidance mandates this. The sum of the projects proposed here should allow others to confirm the hypothesis that Germany Beck is indeed the site of the 1066 battle of Fulford.

This text forms the basis of chapter 7 of the report.

• Because of the context, with the site in danger, the projects could be divided into those that are essential to grab all the information available before the site is destroyed; so called ‘rescue archaeology’. 
• There is another set of vital projects which lie outside the threatened area that might provide more information or test the existing work, but there is no particular urgency for these. 
• Finally there is a set of projects that will further our understanding of the battle using the material gathered.

In the ideal world, such an artificial division of the projects would be unnecessary. 

All of these projects are vital for a proper analysis of this site. So there can be no question of simply ‘rescuing’ the data and then destroying the site. Such an irreplaceable piece of our heritage must be properly investigated before society can be expected to make a decision about the site.

The work would be undertaken in a way that is led by the process of investigation and discovery. It shames us all that we find ourselves having to propose such an artificial programme.

The area north and south of the beck requires a full soil profile. This must be more than the engineering works required before the road is built and must do a proper appraisal of the layers so that the full history of the land can be interpreted.

The work is particularly important as it will allow the shape of this land where the battle was probably fought in 1066 to be mapped.

At the north-east corner of the cemetery there is some uncertainty about the precise shape of the land since the sand in that area is not well stratified, with no ‘marker’ layers. This area provided a secure flank for the English and constrained the Norse force on the southern bank. A more detailed model of the underlying terrain, which is close to the embankment for the proposed access road, would complete the survey of the beck.

The area to the east of Fordlands Road bridge has shown that there is alluvium capped below clay that can be traced towards the east and lies above the charcoal pits. It is important to try to date this layer, possibly with carbon dating since it is the only relatively undisturbed part of the old ford. Permission has been granted by the Secretary of State for this area to be destroyed even though no investigative work has been done on an area that is at the very heart of the battle.

The work to define the shape of the ford beneath the playing field, described in the text as an amphitheatre, can be done as a part of the work needed to understand the local hydrology that produces water-spouts. The air photos and old maps suggest that the base of the ford might have had three flows meeting. This will allow an accurate map of the slope, composition of the terrain and precise configuration, which will improve our understanding of the events at the ford. It is understood that some conduits were built to carry the water but no documents relating to this construction have been found. The image (right) was taken during one of the regular floods and probably points to one of the lost drainage channels. It was pointed out that these will by-pass the flood defences being planned but this helpful advice has been ignored.

A systematic soil survey of the area beyond the eastern boundary of Fulford Cemetery as far as the golf club. This could be done in conjunction with the various geophysical techniques. Many paleo-channels are visible on the LIDAR map and hedges and tracks will help to plan this work. This should reveal how secure a flank-obstacle this was and if it provided a suitable retreat route for the defeated Earl Morcar.

• The results of these surveys will provide useful data for planning or interpreting other work, such as the metal detecting which is discussed later.
• We can use the data to reconstruct the various landscapes, since this is designated in the plan as an archaeological zone. The land surface, since the last ice sheet retreated, could be understood.
• There might be a good case to restore the natural profile of this area and respect the natural contours for the flood ponds, should they be built. In the worst-case scenario, at least this part of the site could be restored to its 1066 look.
• It would be helpful to our understanding of the retreat phase of the battle if this landscape were plotted using the existing trenching and other techniques to identify the old wet areas. Several channels were identified and a second phase of archaeological work by MAP was also revealing.

One possible speculation during the project was that iron salts, which were the product of rust, might show up in the record. This technique has been employed at Riveaulx Abbey to measure the levels of iron extraction. During the XRF work, several random soil samples that were to hand were tested. Some samples showed an abnormal level of iron. These soil samples, it was later confirmed, came from the Ings below a known buried dump of old machinery south east of the old Terry’s chocolate factory. This experiment suggests that a systematic testing for iron in the soil, using XRF or other techniques, would be worth a trial and should be used in conjunction with the soil surveys listed above.

Furthermore, if this does provide a viable method, then it opens the exciting possibility that a number of enhanced areas of iron in the soil can be dated using either some carbon material found in the sample, or using the land build-up model to estimate the date.

If a pattern of enhanced ferrous levels could be found and dated to the century after the battle this would be very close to the ‘smoking gun’ that would remove any sensible doubt about the beck as the locus of the battle in 1066.

The testing of the soil samples for ferrous traces should be a part of the soil survey methodology.

Phosphate Mapping

Several sites in Scandinavia where long-term occupation was suspected have yielded excellent 'maps' of human activity when subject to the sort of phosphate analysis used in agriculture.

Because new techniques will be introduced and existing methods will become more sensitive and accurate in future, the unique nature of this site demands that a systematic matrix of soil samples are taken and stored before the site is disturbed. 

It was obvious during the planning process that the data they employed about future flood levels did not match reality. Leaving aside the question of whether an application based on such erroneous data should not be recalled for reconsideration by the planning authorities, a much more sophisticated hydrological model, that is based on some accurate and original local data, will be needed to allow any project to build houses in this area that will not flood.

In addition to recognising the real level and frequency of flooding, the streams buried below the playing field must be introduced into any model as their effects have often been observed and existing plans have been at fault in ignoring these. (See above)

If a good model is built it might allow a more precise estimate of the timing and height of the tides reaching along Germany Beck to be calculated. This hydrological information will not have a profound effect upon the interpretation of the battle but the understanding of the buried stream is important to our detailed interpretation of the events at the ford.

Denying floodwater access in one place, will force it to overflow elsewhere. There is a real danger that even those parts of the site that are not destroyed by the road will suffer by their proximity to it because of increased flooding caused by the damming of the beck. Only an appropriate model will allow a proper design which might make it possible to prevent this disaster.

We support English Heritage’s suggestion that the peat layer along the line of the proposed access route must be removed and analysed. A mitigation strategy to investigate and possibly remove all of this material already forms part of the planning consent.

The anaerobic conditions that exist within the peat make it possible that organic objects might have survived. Boots, cloth, wood from weapons and fragments of shield might remain to be identified, as well as metal fragments.

It is possible that the top layers of peat, which might have contained some of the battle evidence, has already been removed for fuel since carbon dating suggests the top layer of peat predates the battle by nearly 200 years. But it is equally possible that the growth of peat had already been choked by the influx of alluvial material and this is strongly suggested by the build up of alluvium at the ford. So the peat in this area would be the 1066 surface.

The boots of heavily clad warriors would have penetrated many centuries of the peat layers. So it is not just the surface, but the peat layer could well hold pieces of forensic material, which is why the peat needs to be removed in its entirety for conservation and examination.

Once the area has been searched for metal, two geophysical investigations (magnetic and resistivity) should cover the whole threatened area. The removal of the surface metal, which is discussed next, should improve the image produced by these geophysical techniques. So it is suggested that metal detecting should precede the geophysical surveys.

Now that the archaeology of the battlefield it better understood it is necessary for a new and consistent survey to be conducted to cover the entire area, including areas reserved for flood ponds and parkland.

The aim of this work is to identify any areas where post-battle activity took place. This might help to identify any areas that have been disturbed for the disposal of bodies even if no mortal remains survive. The work might also find possible charcoal pits.

It might also identify the metal working areas where iron concentration lies beyond the range of surface metal detecting because geophysics might give a meaningful signature for hammerscale and slag.

This might help to localise the places where metal working took place, and the location of paleochannels which played a part in the retreat phase of the battle and will aid the mapping of the old marshlands adjacent to Germany Beck.

It is very important that all this is done before the land is disturbed.

Less than a third of the available site has been surveyed for metal and all of this was a preliminary scan. If this is to be the final chance to check the land for metal, given the unique nature of the material identified, the work must be comprehensive and thorough. This might be the only place where such an assemblage of post-battle metal working will be found because of the extraordinary historical setting of 1066.

There is a large hole in the centre of the present study area and this is exactly where we now believe that the battle action happened. So a complete survey of the metal work is vital and this is now mandated as one of the planning conditions before detailed planning permission is granted. There are several key sites (zones 3 and 4 identified in chapter 4) where the surface metal scan will provide a valuable confidence check on the existing data, as some predictions were made.

It is important that this work is undertaken using a consistent working method that will allow comparison with the existing work. It would be unacceptable for this work to be done in a way that did not enable it to be compared with the existing data set so that it can be integrated with the existing work. Using GPS, within the grid established in area 9, it might be possible to identify the ‘point of origin’ for the ironwork.

The land must be well prepared before it is searched. The work needs to be coordinated with the farming calendar ; access after the autumn or spring ploughing and following rolling, with the work spread over two growing seasons, would be the safe minimum. Some areas might also have to be deep ploughed to expose the material.

Having experienced the skilful manipulation of the system by those who want to build over this landscape, the requirement to search all of the land must not allow any exclusions. It would be desirable to impose a condition to frustrate the research work beyond the immediate building zone since there could be agreements with neighbouring land-owners in place.

Some written undertaking from the builders along the lines of the following would be suitable:

"We will seek to encourage the landowners, their agents and trustees to grant permission at times suited to agricultural constraints for investigative work on and around the Germany Beck site [in this example] for any work related to a better understanding of the historic or environmental events of 1066. We will do this in a timely manner that will enable work to be conducted to inform a proper judgement about this land before any final decisions are made about building work provided that this work is also undertaken in a timely and cooperative manner.’

For the planning system to work, it must be provided with good and relevant information. So it cannot be right that those submitting any proposal are allowed to prevent the work of others who seek to provide this information, if they are unwilling to do this themselves and the authorities feel unable to demand the work.

The whole of the area north of the beck is awaiting a metal survey. 

It would be surprising if there are not further reprocessing areas along the north bank, given the number found just across the beck. Indeed, since the publication of the report a member of the public has reported finding material similar to the hearth debris reported on the north bank. This was deemed by the finder to be sufficiently interesting to report to the Yorkshire museum, but sadly the significance was not recognised and after identifying it as hearth debris, it was disposed of.

There are four areas that have been identified as possible reprocessing sites.

The two most important ones in terms of yield lie within the threatened area. A systematic search for metal, including hammerscale and slag, would localise areas where metal working took place. Although not all the identified reprocessing sites lie inside the threatened area, it is vital that all the areas are investigated before any irreversible damage is done to the site. This is because all the reprocessing sites must be compared in order to give a true picture of the work that went on after the battle.

We believe that the finds have been moved on by ploughing and earthworking. It is conjectured that the material has only come to the surface in recent decades, since few of the items have plough marks. If this reasoning is correct, then the source zones of the ferrous material might still be undiscovered and undisturbed.

Once any sources have been located a formal, stratified excavation is needed, with all the spoil being checked for metal, hammerscale and furnace debris.

If the hypothesis suggested by this work is correct, then the possibility exists that below the surface to the south of the beck there is an area covering possibly 5 acres that needs to be subject to this systematic inspection. When the area to the north of the beck is finally metal detected, it is possible that another extensive area with metalworking spots will be revealed. The potential prize from this work will be many different hearths and their associated tools, plus the work in progress.

The point is again made that the evidence so far discovered suggests that the work was abandoned after the destruction of the Norse army at Stamford Bridge and their base at Riccall. If enough material has survived in-situ it will give a unique insight into the metal working techniques of that time.

With over 50 and possibly three times that number of items already found that have been related to metal working, there could be very much more to be found. The possibility exists that an extensive area of metalworking exists on this land. There is a chance that this is one of the rare archaeological treasures which some unique event has ‘frozen’ in time. The size of the prize more than justifies the effort required to uncover it.

Most of the finds would benefit from further conservation. The use of x-rays has allowed many of the objects to be identified but only two of the smaller objects (arrowheads) from the reprocessing areas have been conserved. The material is stable but work will be needed before the items can go on display.

There is a set of finds from area 7 that have not yet been subject to any detailed analysis, x-ray and photography, since they were given to MAP after the survey of the spoil heaps. Data from this material needs to be included within the existing as well as the future work.

The long-term requirements for conservation cannot be assessed until the area has been fully investigated.

A start was made with the statistical analysis. Once the survey work of all the areas is completed, work on the finds data can be continued in the expectation that it will yield information for the benefit of future researchers. This analysis will be much easier if there is consultation and planning to ensure a consistent method of study. The successes as well as the errors and failures of the metal work undertaken should provide a good basis to design an efficient and effective method for recovering ferrous finds elsewhere.

There is much information to be extracted from the ferrous material, including the morphological analysis introduced earlier. This is a manpower intensive activity if it is to use the objects themselves rather than the images, but might provide a usable method for future researchers if the whole collection is analysed in this way.

When the full data on horseshoes is available from the battle area and surrounding fields, it is hoped is to carry out analysis by shape, quantity, mass and find-density over the whole area. There is more information to be extracted from the horseshoes. The main relevance might be to point to land use as there is nothing yet that allows any suggestion that mounted warfare played a part in the battle at Fulford.

Some carbon that can be confidently associated with the hearth material might be sought. There might be traces of carbon trapped within the hearth debris and this might serve to provide a ‘not later than’ date – a ‘not before’ date from subsequent centuries would mean that the material was not relevant to the battle.

It was felt appropriate to wait until the whole area had been searched before the various techniques were undertaken to examine the crystal structure of the iron. This can be done by taking thin sections or cutting and polishing a surface. The patterns can reveal much about the working methods employed to forge the metal.

A micro-metallurgical analysis of the crystal structure of the tools, anvils and billets as well as the hearth bottoms will help confirm their purpose because iron can exhibit patterns of use at the crystal level. This work might help to identify the true nature of these finds as metal that has been worked (beaten) has a different structure from annealed (heat treated) iron.

In December 2009, Oxford Instruments kindly loaned a portable XRF machine along with an operator: The University of York, Department of Archaeology agreed to provide another technician and one of their laboratories so that we could safely conduct the measurements. I would like again to express my gratitude for their generous gift of resources, time and expertise. The work lasted two days during which 180 readings were taken and 120 items tested.

All the items identified as related to the metal reprocessing areas had no element that one associates with modern alloys. So we cannot say the items are ancient but we can say that they are probably pre-industrial.

As is so typical of research, the results answered one set of questions and opened more lines of inquiry.

There appeared to be patterns of trace elements among the finds and, once the complete assemblage has been tested, it might be possible to associate items that were not found in the reprocessing areas with the ferrous material from the hearth areas. At the time of writing it is not certain that a statistically significant ‘fingerprint’ for the different metals can be identified as the data set might be a bit too small.

However, if this does prove viable, it will make sense for all the iron debris scattered over the field to be tested. This might allow an unprecedented level of understanding of the battle itself. The significance of this bears repeating: It might prove possible to relate the fragments scattered round the site to the metal found near the hearths because of the XRF signature. It has never been attempted before, but the ‘orphaned’ item of iron could be related to the battle if they match material at the processing sites.

So the archaeological planning for this site should test this hypothesis and then be prepared to conduct a 100% test on all of the iron so that each piece can be plotted across the site.

The recommended strategy for researching metal working sites is to look for hammerscale. Indications from the original work suggest that there are ‘phantom’ fragments of iron in areas 10 and 9; a technique for trapping and mapping this material would be a challenge.

Once the metalworking and geophysical work has been assessed, for any area where surface evidence of reprocessing work has been identified, a programme of excavation might be required. The details of what is needed can only be determined when the pattern of metalwork is identified.

In the trench 5 area from the pond excavations by MAP the limited detecting work on the spoil revealed a high ratio of ferrous to nonferrous, which is suggestive of the proportions of iron associated with the hearths. The density of finds, i.e. the amount found per unit area, was over twice that of the next best trench and similar to the levels found around the other hearths.

It has been noted that this excavation was close to row L & M in area nine (zone 4) where a significant quantity of billets, tools and a small hearth bottom were found.

This invites several questions.

Are the high ferrous proportions an indication that more ferrous material lies below the surface in this field?

Is the nonferrous material that we found on the surface mainly modern scatter with ‘older’ iron lying just below?

And does this reflect the rate of deposition or the agricultural activity in this field or has some process buried the iron and concentrated the non-ferrous items nearer the surface?

A re-examination of the data and a chance to inspect the material from this area, which is in the possession of MAP, would be a useful preparation to any other excavation work on the field.

At least one of the areas beside Germany Beck where a charcoal stain was identified needs to be excavated. A test trench should show if the whole area of the pit needs to be explored. If one pit can be dated to the time of the battle, using carbon dating, then the other areas need to be subject to some inspection.

Field walking to try to identify more stone hones, hearth material and slag is necessary. The existing fragments were all found in close proximity to the smithing areas or hearths. Such work could go alongside the metal detecting since the conditions for the one also favour the other.

The identification of several water vole colonies along Germany Beck calls into question the veracity of the work presented so far as a part of the planning process. A new survey is needed for the ecology of the area. While this is primarily significant to the planning process, a proper study of the ecology will help to appreciate the role of the beck as the drain for the hinterland and this is relevant to the battle and hydrology.

The original desktop study for Germany Beck noted some hedges that might have been there at the time of the battle. Hedges have an obvious relevance to the battle landscape.

In particular, our own work confirms the earlier study, that the hedges running along Germany Lane could be 1000 years old.

There is much more work that can be done and techniques such as carbon dating and pollen analysis. It was noted in chapter 3 that arguably the oldest hedge was removed, but traces of it have survived and are re-growing. It should be possible, with some sample trenching or core sampling, to identify the species and the date of this hedge.

Serious consideration must be given to its restoration as a way to stabilise the land and provide the various habitats that make this beck such a valuable wildlife conduit to Heslington Common and beyond. The findings here might also lead to a modification of any plans to remove any other existing hedges.

Pollen and heavy-metal analysis of the stored core samples already taken could help to clarify some outstanding questions about the landscape.

The different methods of recording meant that normalising the data for analysis was a challenge. Indeed, the time that would have been required to convert the blocks to notional GPS readings or to allocate those finds with a GPS grid to a notional block was abandoned as being too time consuming. Given enough time and computer software, the two data sets could be combined. It was a shame that the GPS was introduced but the data is available should anybody have the resources to interpolate it. There is more data to be analysed than is presented here. Fortunately the later metal surveys were not at the heart of the battle site.

The morphological analysis that has been attempted on the ferrous finds adopts a statistical approach to give some relevance to the fragments thrown up by centuries of ploughing. The initial results will, I hope, encourage an investigation of this technique as another possible way to identify the landscape of battlefields.

Much effort has been expended in the search for the two roads that are believed to have run south of York. The locations are strongly suspected and it is possible that little remains of these minor roads in the area of the battle site where flooding has either buried them or caused the routes to be relocated.

However, working further back from the fording places, it might be possible to project the routes and this would help to clarify the approach routes of the armies. It would be an excellent project for community archaeology groups.

No work was undertaken at Riccall as a part of this project. It would also be very interesting to identify the remnants of the fleet of 400+ longships.

Norse and English literature suggest that the invaders were almost wiped out and this would imply that the battle we know as Stamford Bridge encompassed the base at Riccall. If fire took hold then most of the ships might have been destroyed, consequently there might be some significant remains provided flooding buried them quickly.

The possibility of some dendrochronology and/or carbon dating on the some of the trees fringing the beck as it crosses the Ings would help to confirm the interpretation of the landscape and identify those trees that stood in 1066.

There is a possibility that one of the ‘mother trees’ might be identified that has long since been buried by the rising level of the Ings. This work would provide important confirmation of the line of the beck in earlier times. No parallel situation has been identified so it is impossible to know if this would work. It is possible that the lode tree has rotted removing any datable remains.

A series of boreholes could be drilled parallel to the Ouse where Germany Beck used to enter the river. This would help to confirm the interpretation of the battlesite as this is where the right flank of Morcar’s army was turned. These bores would need to be at least 3 metres deep.

Existing work has indicated that the bank is not well stratified, with little variation in the colour of the alluvial material. Therefore it might prove necessary to dig a trench where any subtle variations in colour or texture might be visible. This would be deep, near to the river and so it is technically challenging.

Post Script Contractors for Yorkshire water laid two, 1 meter waste water pipes across this area. A programme of work was agreed but was not adhered to citing safety issues. A request has been submitted for this work to be done as a part of the land restoration work being undertaken by the contractors.

The ‘charcoal abundance score’ in the Ings that was discussed in chapter 2  of the Finding Fulford report showed more charcoal than elsewhere in the profile (except at 660 mm) at depths which correspond to the dates of 1060-1080. More work on this might provide data for the immediate pre and post conquest period.

Further investigations will be required to determine if this charcoal is a residue from the battle or from the harrowing of the north three years after the battle. If grazing animals were killed, then it would be necessary to burn the land to prevent woody species becoming established.

The decision was made not to carbon-date any of these samples because it was not possible to determine if the pieces recovered in a core were representative.

Carbon dating is expensive and it is important to ensure that any piece is representative of the assemblage. If the sample happened to be a piece of coal or from a very old tree, the dates produced would not provide any useful information and could be misleading. A plan was proposed to excavate the area and identify a sample of charcoal from a twig as this was most likely to be contemporary with the activity which produced the charcoal. Uncertainty about the ownership and access rights of this section of land has so far prevented this work but it provides an exciting project for the future.

As well as carbon dating, a study of the dendrochronolgy and taxa is emerging. The charcoal might have an important contribution to make to the Fulford investigation.

The data obtained from the cores needs to be incorporated into a 3D terrain modeling system. This will improve the interpretation that is possible about who could see what on the battlefield. A physical model was built.

These are the tasks and investigations which will allow for a proper interpretation of this battle. However, there must be an awareness of emerging technologies which will help to interpret this site.

This emphasises the dilemma often faced when sites with heritage value need protection. We are learning so fast and improving techniques that even when a site cannot be ‘proved’ using existing methods, a decade later such proof might be forthcoming. But the planning guidance recognises this and unambiguously and repeatedly says that sites should not be disturbed.

In the event that final permission is granted to build the access road it is vital that all this work is done to gain the maximum amount of information before such a cultural crime is committed.