FIELD WALKING
PROTOCOLS
INTRODUCTION
There are many ways of conducting a field walking survey and protocols are available on line from web sites in many parts of England. Several counties have their own recommended protocol for surveys carried out in the county and if it is at all possible one of these should be followed. However, our experience has shown that different groups will have different objectives and requirements and these lead to necessary variations in procedure. The methodology for field walking followed by BHTA is based on one devised for BHTA by Jenny Brown and Peter Incker of the Trent & Peak Archaeology. Theirs was a modification of the methodology that would be followed by professional archaeologists on a commercial survey and is outlined in a report they produced for BHTA. During the first and subsequent seasons it became evident that further modifications were required to take account of the special requirements of amateur groups and the external conditions under which BHTA was working. Here is presented the methodology as finally developed. It is meant to be used only as a guide for amateur groups who may wish to embark on projects like this.
BEFORE THE PROJECT BEGINS
Field walking on a parish-wide basis is a big venture. In our case it involved walking around 870 hectares and collecting over 50,000 finds. It took 74 weeks of field walking. There are no professional archaeologists among the membership of BHTA. This meant that to identify the finds BHTA would have to bring in specialists. It was decided, therefore, that the project could only be undertaken with external funding and an application was made for a grant from the Local Heritage Initiative to fund it.
Before making the decision to do such a project and putting in an application for a grant there are several important steps that have to be taken:
- Calculate the area of land that is to be field walked.
- Consent from all landowners and tenant farmers is necessary. It cannot be assumed that all landowners will be sympathetic to a project like this.
- Ensure that there is a sufficiently large pool of volunteers to do the work. In our case this meant having a pool of over 40 members available of whom about half would field walk in any one week. With this manpower resource we field walked between 11 and 12 hectares per week.
- Identify the specialists who will identify the finds. At this point the decision has to be taken whether to collect finds from all periods of be restrictive and stop at, say, the Medieval period. We decided that using diverse volunteers for the project it was safer to collect finds of all ages rather than risk not picking up important material through inexperience.
- Establish that there is a member who is prepared to use their property as a headquarters where the kit and collected finds can be stored. 50,000 boxed finds fills a single garage. It is also useful if the finds can be washed and processed at this same locality.
- Confirm that there are members of the group who can do the various key tasks required. These are someone to organise the fieldwork, someone to take charge of the finds processing, and someone to be responsible for all aspects of computing. A parish-wide project is too big for these to be done by one person.
- Decide on the type of survey; that is whether you want to do a 10%, 20% or any other variation. Before we started we discussed these options with people who had done similar surveys and were told that for a general, reconnaissance survey it did not matter if it was 10% or 20%. In reality, when the Lottery funding agencies give a grant they give a time limit for the duration of the project and this will define the type of survey. In our case we had to do a 10% survey so as to finish it in the prescribed time; that is 2 metre-wide transects 20 metres apart. Commonly, in a 20-metre survey, stints are set at 20 metres, effectively creating a 20 metre grid for the field. We chose to have 5 metre stints because of the greater accuracy this gives in plotting locations.
Once the decision to do a project is taken it has to be costed. Although there will be considerable local differences below are some of the costs incurred in our project as a guide.
- Materials and equipment £1200
- Professional fees £17000
These vary greatly and it is worthwhile discussing the day rate with each specialist before commissioning them and agreeing a number of days for the job. - Publication costs £6000
These depend on how you aim to publish your findings. In our case there is a book, a web site and an exhibition - Office costs £700
SPECIAL REQUIREMENTS
In a commercial survey professional staff would work full days and the same team would come out every day. When volunteers are used, totally different conditions apply. In the BHTA survey there were eleven sessions of three hours each week. They began at 9.30 am and 12.30 pm on Monday to Friday and finished for the day at 3.30 pm because of the poor winter light after that time. On Saturday there was only a morning session. Although there was a physical handover between sessions during the day there was no continuity in the workforce between sessions or days and as many as 45 different individuals may have been on call during the average week. Effectively this meant that there were eleven different teams doing the work. Many of the individuals concerned were doing only one session a week; some less than that. The procedure followed had to take these factors into account.
PERMISSIONS
Before field walking can be undertaken permission has to be acquired from the necessary authorities. In the case of Bingham all farmland is owned by the Crown Estates and farmed by tenant farmers. Permission from both was required. Once permission is acquired it is important to discuss crop rotation with each the farmers. There is considerable variation in the way this is done and it can limit severely the availability of fields in any one season.
INSURANCE
The Crown Estates insisted that BHTA was covered by public indemnity insurance for the duration of the project. This was acquired through the Council for British Archaeology, to which BHTA became affiliated. The insurers asked that each field walker signed a waiver stating that they had been warned of the dangers of field walking and that they agreed to do it at their own risk. (Attachment A)
KIT
The basic requirements were:
- 50 metre lengths of polypropylene rope, knotted every five metres and with lengths of red and white barrier tape threaded through the knots. Each rope was attached to a plastic reel obtained from a hardware shop where they had been used for holding chain. A long bolt was fitted to one side of the reel to act as a handle for winding the rope up. A cone was used as a spindle. At each end of the rope were attached metal rings about 3cm in diameter.
- Flags mounted on four-foot canes.
- Survey pins. The ones made by BHTA were redundant carbon fibre and aluminium archery arrows with corks fixed on the top and painted white. A length of barrier tape was attached just below the cork.
- Two sets of Perspex labels with letters painted on them, one in black the other in white.
- Wooden boards mounted on spikes that could be pushed into the ground. On them were painted sequential numbers marking the stints. The sequence was 11, 21, 31, 41, 51, 61 etc.
- Metal tent pegs used for attaching the ropes to the ground.
- A 100-metre measuring tape.
- Optical square measuring device.
- Small white flags made using carpet tape and 20cm-long sticks for marking the location of finds.
- Finds bags. The most commonly used were 3.5 by 4.5 inch plastic bags.
- Clip boards with a ballpoint pen attached to the board by Velcro and a bulldog clip for holding finds bags.
- An A4 folder to hold guidance notes, the session sheets (Attachment B) and copies of the waiver to be used by new walkers (Attachment A).
- Three survey stadia.
- A notebook for recording all information about each field.
PROCEDURE IN THE FIELD
Baseline
Where a field was not rectilinear in shape the baseline was set across the widest part so that all transects would intersect it. The pole position was where the end of the baseline and the first transect intersected. The baseline and the first transect were set at right angles to each other using an optical square. An attempt was made to set the first transects parallel to a hedgerow and crop lines. At the start of work in a new field the first six transects were set to the first 50 metres. Survey markers were left in place on the baseline and at the first 50 metres until the six transects were completed. Then the next six transects were measured in using the markers on the first six as guides. When the baseline had been extended across the whole field the position of the first and last transects on it was measured with respect to topographical features shown on a map. A sketch of the field with this information was made in the field notebook.
Transects
The transects were identified by letter, starting at A. In wide fields when the whole alphabet was used up the lettering continued with AA, BB etc. Each transect was divided into stints, in our case at 5 metre intervals. These were numbered from the baseline sequentially from 1. Where there was sufficient space behind the baseline for the transects to be continued backwards the stints were numbered 199, 198 etc back from the baseline.
Field walking
50 metre lengths of transect were walked at a time. The start point of each transect on the baseline was marked with a flag and a perspex square with the letter painted in white on it. Another flag (any colour will do) was placed at the first 50 metres. The white letter was left at the baseline until the survey was completed. A second letter, usually the black one, was carried by the field walker. At the end of a session it was left at the end of the portion of a transect that had been surveyed as an indicator to the person who was to take over the transect at a later session. On completing the first 50 metres the rope was laid out for the next 50 metres in line with the first. This can be done by eye. It is worth measuring the ropes at intervals through the season because even polypropylene varies in length according to humidity. As long as variations in the lengths of individual ropes are allowed for no further measuring is usually necessary except on particularly long fields. Walkers should proceed at about the same speed. The end of each 50 metres surveyed is marked with a survey marker for sighting purposes when extending the line. They can be left in place until the transect is completed. Stint marker boards are usually placed at each 50 metre marker on Transect A and can be left in place until the first six transects have been completed.
Collecting finds
Finds were collected from a two-metre wide strip along the rope, one metre on either side. The position of finds is located with a small flag. On completing a 50-metre section of a transect the finds are re-examined before putting into bags. All finds in one stint are bagged together.
Numbering
Each bag was given a number consisting of the following components: the field number, transect letter and stint number and the initials of the field walker. The field number can be derived from the OS grid reference for a point in the field by removing the first digit each from the easting and northing in a six-figure OS grid reference; for example SK 704 387 becomes 0487.
Transect spacing
In this survey 20 metres spacing between transects was adopted.
Completion of a set of transects
On finishing a set of six transects all markers are removed except for those on the base line and at the first 50 metres and along the last transect to be walked. These are used as guides when setting the next set of transects. The stint board markers should be moved across to the last transect.
End of session
At the end of each session all finds were collected up, the bag numbers checked and they were put into large bags, which were marked with the field number, date, morning or afternoon session and the transects done for delivery to the HQ.
Information about the session was filled in on the session sheet; i.e. names of the field walkers, what they had done and so on (Attachment B). In addition, instructions were written about the session for the benefit of the next group of walkers to work in the field. After the morning session all necessary survey markers, flags, stint boards and so on were left in place. At the end of the day all items that were not necessary to guide the next set of field walkers were removed. This is necessary because some farmers ask for it to be done, but also so as not to attract attention of “Nighthawk” metal detectorists, who have been known to go over fields where they spotted field walking activity.
End of the day
At the end of each day the kit was delivered to the base and stored for the night. The clipboards need topping up with new bags and checked to make sure each board has a pen that works. An eye needs keeping on the supplies of bags so more can be ordered as and when needed. The session sheets for the day are collected and new ones put in the folder for the next day.
Crops
The best conditions for field walking are a few weeks after ploughing or drilling, when the soil has weathered sufficiently to reveal the finds on the surface. Most farmers take a relaxed attitude to field walking, but care should be taken even with them not to go on wet ground and puddle the soil on clay ground. Care should be taken not to leave boxes or unused kit lying on the crops. They should be placed under hedges or on tracks.
Completion of a field
When a field is finished the first task to be done is to measure the position of the first and last transects where they intersect the baseline. Then, all kit and markers should be removed. It was found in the first season that the cork survey markers were easily overlooked and often left behind. For this reason lengths of barrier tape were attached to the tops to increase their visibility.
Responsibilities and organisation of field work
During the whole project all the baselines were set out by one of two people to minimise variation in practice. It was their responsibility to see that each field was set up before the walkers started work on it.
For each session the person designated to pick up or return the kit to base was nominated as in charge and took responsibility for all aspects of work in that session.
Each week of work was planned ahead. An e-mail was sent out by the project organiser each Thursday evening to all field walkers asking them to declare before Sunday evening when they could come out during the following week. Those not on e-mail were called by telephone. The rota for the week was closed on Sunday evening. The minimum number of people allowed to go out in any one session was two. On Sunday evening a copy of the rota was sent by e-mail to each person designated to pick up the kit. A copy was also sent to the individual organising the finds processing and one was put into the A4 field folder.
New field walkers joined throughout the length of the project. Each one was trained by one of a small number of experienced field walkers to ensure that common practice was sustained.
PROCESSING FINDS
At the end of the day all finds were delivered to the headquarters were they were processed centrally by teams dedicated to the various tasks. The tasks carried out here are given below.
The day’s finds
These need sorting for each session (preferable not to mix sessions) in order and laying out to dry in boxes. (Boxes used were vegetable and fruit boxes obtained from Bingham market.) Depending on the number of walkers and number of sessions in a day there may be up to five boxes of finds, which can take anything up to an hour and a half to lay out.
Washing the finds
It takes 20/30 minutes to wash a box of finds. Washing was arranged in three or four sessions a week with teams of up to three at a time. You need a minimum of one washer/session for every six walkers/session. The work is fairly tedious once the novelty wears off but has to be done methodically so nothing is missed.
Bagging the finds
This is done once the washed finds have dried. The length of time it takes to bag a box depends on the number of different types of find per stint. One bag is used for each find and all the details of field, stint and transect number and field walkers initials have to be transferred to each bag.
Categorising finds
After bagging up, the finds are roughly classified into an agreed set of categories, (such as flint, pottery, brick or tile, metal, bone, shell etc). The pottery is then divided into agreed classes (such as Iron Age or Anglo-Saxon, Roman, Medieval, post medieval, etc). The practice here can vary depending on the level of knowledge of the person doing it. In Bingham, it was a two-stage process. At the first sorting stage all “old” pottery, that is up to and including early post medieval, was put together. As volunteers gained experience this group was further subdivided.
Recording
Database entry requires two people. A Microsoft Access relational database was used. The fields for detailed information in the database are left empty. With practice you can do about 80 per hour on a good day provided they have been bagged correctly. You can only do it for about 2.5 hours at a time before the tedium leads to errors. The information entered at this stage is:
field number
stint number
collector
find category
At this stage each find is automatically given a unique number, which is written in red on the find bag. Data entry was done one field and one category at a time so that, as far as possible, each category for any field had a sequential set of unique numbers.
Plotting locations
Because BHTA did not have access to GPS or an EDM, grid references for the locations of finds had to be obtained from scale maps of the fields surveyed. Two methods were used.
The first method relied on making a scale drawing of the transects and stints in a field on a map at the scale of 1:2,500. Each find location was marked on the scale drawing. A transparent grid was used by overlaying in on the map and reading off the grid reference of the finds localities. Eastings and northings were listed in separate columns on a copy of the relevant pages in the database. The grid references were then entered into the database from the paper copy of the table. Done this way the accuracy of the grid references is checked twice. However, this method was laborious and prone to error.
When ArcGIS9 was obtained during the second season a new method was devised. The grid references for the position of the first and last transects on the baseline were obtained on a scale drawing at 1:2500 using measurements acquired in the field. These two points were then plotted on an image of the field in ArcGIS9 and a 20 metre by 5 metres grid devised to replicate the pattern of stints and transects for the field. Grid references were then obtained directly from the screen using the referencing facility provided by ArcGIS9. Two people and two computers were required for this job. One person locates the find locality on a screen showing the grid on the field being done. The cursor is placed at the mid way position between two stint lines. The grid references is read off and the other person enters the information into the database on another computer. A copy of the Ordnance Survey digital 1:10,000 topographic map is essential for this to be done. These are licensed annually and can be obtained from any OS licensee.
Numbering
Individual finds are numbered using the five digit unique number in the database. They are marked in black or white ink directly on the find. This is done after database entry is complete. Numbering was done ideally in teams of three; one person takes them out of the bags and lays out the finds and, after numbering, bags them up again; two people do the numbering. This process offers an opportunity to re-examine the finds and check that they have been classified correctly.
Identification
Finds were sent to specialists for identification only after they had been numbered. Once they were identified and described the information on them was entered into subtables in the database. One subtable was set up for each of the classes of find that were used, such as Roman pottery, Saxo-Norman pottery, Medieval pottery, glass, clay pipes etc.