Metal Detecting in the USA

Elliot removing layers at the field slave site

Continued from previous post…. Can this program get any better?  Even though a few of the detectorists are nursing sore muscles, moaning at the early mornings and trying to figure out how to plan apparel for 28F mornings and 60F afternoons, we are rapidly getting into archaeological rhythm.  (We’re only half-joking when we say the Arlington house needs a candy dish full of Motrin!)

On Thursday the detector stayed in the house and I spent some of the day working in the lab and on the archaeological site, learning how the finds are dug, screened, cataloged and interpreted.  I am also humbled by how hard this work really is.

This site is at the base of Chicken Mountain and was initially identified by metal detector survey which, of course, allows a large area to be covered quickly – but only indicated metallic objects. Based on the scatter of metal, the team subsequently excavated a shovel test-pit survey on a 10-foot grid, and the data was compared.  The two patterns correlated quite well, further endorsing the viability of metal detecting for site definition activities.

Our windswept dig site

With Shovel Test Pits (STPs) soil is sifted and the artifacts recovered, leading to a decision process about excavation.  According to Matt, it took about 3 days for the metal detector survey, and three weeks to cover the same area with the shovel test-pits. It seems to me that embracing skilled detectorists during site surveys can reap very tangible rewards, and save a lot of time.  The correlation between the STP data and the detector data become very clear when you look at it on the lab’s heatmap software.  When I saw this, I felt suddenly quite useful to helping to document the history of our country, one nail at a time.

So, with my shiny new trowel and well-medicated muscles, I spent time at this site Thursday.  The units were positioned based on evidence, and extended in a checkerboard style as long as artifacts are discovered.  Rich deposits may warrant a full excavation of all squares, resulting in the entire area having layers removed, screened and inspected.

Elliot patiently showed me the methodology, also pointing out that breaking a survey string means you must buy a drink for your unit-mate and collapsing a unit wall meant a round for the entire team.  As I hobbled and stumbled through the site, I was sure to come out of this little experiment a bit poorer.  Aside from an embarrassing head whack on the sifting roof (“pwang!”) I came out unscathed.   (Later on, my friend Jim showed the team how to use the Garrett pinpointer to scan the archaeological units between scrapes, showing yet another transfer of knowledge.)

This thin copper coin was found at the site.

I spent some time removing layers from one unit and it takes some getting used to.  You must find ways to make the trowel work for you, not against you.  Elliot was ever encouraging but I’m sure my half of the unit looked like it was dug by a grub-hunting skunk.  I noticed he had two buckets for every one of mine.  Perhaps I should stick to detecting?

Stick and Mud Servant’s Huts

….The quarters for the field slaves were likely built of logs, with clay floors, stick-and-mud chimneys, and simple wooden shutters. Archaeological evidence suggests that, aside from nails and door hardware, the Madisons provided little resources for slaves to build these simple structures. The homes of the field slaves stand in marked contrast to the house slaves’ homes directly in sight of the Montpelier mansion….
from: Montpelier’s Intro to Archaeology Guide (PDF)

Dolly Madison’s China Hand-Me-Downs

But after a few bucket loads, Elliot and I found pieces of blue Madison era pottery on the screen – very cool.  Often the found china was from Dolly Madison’s collection, after it was possibly chipped and given to the slaves (or pulled by them from trash bound for the dump.)  Besides not having your freedom and potentially being split from your family at any time, I cannot imagine living in a drafty handmade shack on this hill.  It had a steady gale and we, with our high-tech under-armor and layers still came away with wind burns after just a few hours of work.

Next:  Wrapping up and looking forward

Day 3 and 4 flew by… and I barely had time to blog tonight.  I had been looking forward to the Chicken Mountain wooded hunt during the MACP program.   The Chicken Mountain site is one of the least surveyed areas thus far around Montpelier, so the potential for Madison-Era finds was quite real (as I soon found out!)

Chicken Mountain Offers Up Something Shiny

We gathered our gear at the base and spread out into the assigned Grids.  Jeanne and I were again at the top of the mountain for the first half of the day and hitched a ride abord the 4-wheeler and worked downhill.

The hunting is challenging on the slopes but after quite a few high-tones in all metal (iron) I soon had a rather peculiar low tone on the E-Trac a 10-12.  I told Jeanne that it was a special signal  - it sounded like a gold signal or more likely a nickle.  We’d not heard any other low-tones on the mountains other than the odd high-low signal from the shotgun shells.

After pinpointing, I stooped to bend and was amazed to find a late 1700s-1820s “Britannia” button beneath a root.  Later, in the lab, I pulled the artifact bag and gave it a careful toothbrush clean to reveal a shiny, georgous (and rare) Madison-Era button (images below.)

Horseshoes, wedges, roundballs, 3-ringers and other finds were located and detectorists enjoyed the rigorous day.  Some discovered clusters of nails and a jew harp which may indicate servants’ quarters.

The Chasm is Closing

Sitting around having lunch together in the woods was a great experience.  We teased and joked about secretly turning on discrimination and hauling ass through the iron-ridden woods.  The banter increased as the week went on and we all got to know each other.

The abundance of modern, extruded wire was a constant nag I know many of us would have loved to have “programmed-out” on the detector.  After hearing it 20 times it was easy to tell when you’d found more of it, but we still dug every little 1/2″ piece.  Grueling but necessary.

It makes me wonder though.  To truly exploit the efficiency that modern detectors and operators have to offer, someday, archaeology team will need to trust the power of the machines with a skilled operator beyond the binary capability of “beep for metal” especially when trying to locate sites efficiently.

I understand that data integrity is paramount and trusting amateurs’ interpretation of signals is riskier than a “dig everything” edict.   The archaeologists must have sufficient confidence in the data to report on it – sometimes a decision to do detailed work at a site means spending a substantial amount of a projects’ budget!   In the end, sloppy work during the survey can be very expensive in several ways.

So… I’m learning a lot.   A couple of significant observations come to mind.

Give detectorists some insights into what you’re trying to learn and they will get on board – rather quickly.  We can be team players!   Knowing the goals of establishing site boundaries on Chicken Mountain made a big difference in motivation and the feeling of teamwork.   That iron object means something – even if it’s just slag.  It defines the edges of a site – and then when we look at the site in aggregate (as I did later at the computer,) we can build a heatmap to possibly prove a hypothesis – or at least advance the state of knowledge.  That leads to more detailed analysis, and eventually a reconstruction of history the entire world can absorb (e.g. rebuilt Montpelier to the Madison Era specifications.)   The hard work leads to a new, accurate experience for visitors looking to explore the country’s past.

Archaeologists are passionate about not only the work they’re doing today, but what future archaeologists may want to do.  Re-burying iron artifacts may sound crazy to detectorists until you recognize that these objects deteriorate as soon as they’re removed from the soil.  And there is a backlog for proper conservation and interpretation.   Archaeologists essentially think on an infinite timescale – just because they don’t do a dig on a site this year doesn’t mean it won’t be done 30 years later.  They all feel a responsibility to the dataset beyond current activity to a level that we as detectorists simply don’t need to consider.   So.. to archaeologists, just because an object isn’t being dug today doesn’t mean it’s being left to rot – rather – it’s being left in a stable environment until it can be properly recovered, conserved and interpreted.   This is a very unsatisfying point of view to a relic hunter anxious to hold history in their hand on a weekend hunt after working hard all week at a job.

While there is much emerging cooperation and agreement, I can sense the  tension between our two paths.  I put this on the to-do list and feel confident that with the right people talking we’ll overcome these issues.  Conversations sometimes get quiet when the detectorists are talking about finds and recent hunts… and it’s obvious that some archaeologists feel uncomfortable.  It stands to reason.   We did remove objects from their provenance, and that is, according to one archaeologist here, a “very sad thing.”  Another frustration is that some of the more experienced detectorists feel we could help with efficiency more than we’re currently allowed.  We would not be at 100%… but perhaps at a high enough reliability to meet the goals of the site survey?  How would you know that you can rest your reputation on a detectorists’ comments about the yet-undug signal?  There’s some distance to go for sure, but this week has been a big step forward.

Next:  Day 4 – excavations of field slave quarters.

After an enjoyable day working the front lawn at Montpelier on the first day, day two at Montpelier started out with a downpour.  It rained hard and we spent several hours in the archaeology lab hearing Dr. Reeves discuss metal detecting techniques (woodland grids) as well as a fascinating (yes, really!) presentation on dating a site with nails.  We discussed hand-forged, machine cut, and wire nails nails of various types…I also know the meaning now for “Deader than a Doornail”

It turns out that the evolution of nails corresponds quite neatly with the different phases of Montpelier’s history.  They serve as indicators about human activity and age, and there is no shortage!

The rain subsided and we gathered up our gear for a venture onto Chicken Mountain.  Only 4x4s would make it to the hunt spot (my car had to sit this one out) and a 4-Wheeler took us to the top of the mountain.  Jeanne and I began work on our grid square and worked our way downhill. As usual, each found nail is carefully cataloged along with its grid.  While I have always used square nails as a rough indicator of a sites’ age, I will never look at a nail the same way!

Illustration courtesy of Montpelier Archaeology/Dr. Matthew Reeves

Again, working in all-metal mode, we dug all hits – which for the first hour were modern wire apparently related to early logging operations in the area.

Before long I had a pouch full of this wire!  Working downslope, I began to find cut nails just before we quit.  Detectorists further down the hill were having better luck – bullets, suspender clips and other items were turning up.  Tomorrow we hit the woods for a longer session and I have high hopes.

As usual, detecting in the woods is a challenge, especially as you try to cover the area within a gridded pattern.  In some cases there was no hope – downed trees and brush blocked the way.  We took our time and worked as a team – covering several large blocks of the matrix.

Jeanne marks hits along our designated grid square.

At one point I joked with the others that I was going to secretly turn on my discriminator and start hunting copper and brass!   But Jean reminded me that no data is still data, and by covering the grid, even with no hits, we are establishing the edges of a boundary.

Archaeology takes someone who can look at a site from a broader, multi-dimensional point of view – through time, and through data.  The data accumulates and statistical meaning can begin to be extracted.  The first square may have no hits, but the next one might have 5, then 10 and before long we are zeroing in on a new site that needs attention.

This method of zeroing in is also used by detectorists, but usually with far less rigor or precision.  When searching a farm field for a house site, we’ll switch to all-metal mode and begin to mentally record hit density.  Once it reaches a certain level, we’ll switch discrimination on and begin digging good targets.

Illustration from

More to come from day 3.

After much preparation and a 7 hour drive yesterday, today started early with a 7:30 team meeting to discuss previous work and goals of the Montpelier Metal Detecting Archaeology program.   The team gathered around coffee and donuts while Matt Reeves, PhD, director of Montpelier’s archaeological program, introduced us to the team as well as Minelab’s representatives..

We toured the mansion and then sat down in the front lawn to discuss the 10-foot grid that was positioned by the staff.  It struck me as a very positive move to have detectorists and archaeologists sitting in a circle, laughing and enjoying lunch together.

Gary Schafer, General Manager & President of the America’s Region for Minelab, was especially helpful – and I was very happy to talk to him, even showing him the Maprika setup for search and the potential for using Android as the operating system for metal detectorists.  The swag was pretty nice also!

Soon we paired off with the staff to begin the survey, and I began working with Jeanne, an archaeologist at Montpelier.  She took notes and cataloged finds while I learned the process of skewering signals, one for non-ferrous and two for ferrous signals.   At first, it was hard to break old habits of plug size, depth and delay gratification until we had an entire 10×10 square detected thoroughly, in all-metal mode.

I’ve not spent much time with my E-Trac in all-metal mode, so it took a bit of time to get into the groove.  Also, the soil in VA is high in iron concentration, so there were may false signals using my normal programs.   Lance Crosby, the staff detecting technician was a huge help.  On his advice, I dropped sensitivity to 10 and dug between iron signals, leading to more productivity.  My usual setup, it turns out was just too strong for the shallow objects at Montpelier.

So, what did we find?  Nails!  At least 5 varieties of hand-wrought nails and a few other iron pieces.  Each object, iron or not, was considered important and was carefully cataloged by Jeanne from the 10×10 unit we were sweeping.  I think we made it through 5 squares in 3 hours, to give you a sense of the speed.  S-L-O-W.

So, by the standards of a private property coinshooting trip, this was not the best hunt.  Most of the iron signals I heard would have been skipped if I were out on a weekend outing at, say, an old farmhouse.

But in this case you must consider the broader narrative they’re trying to construct.  This was especially well illustrated using the finds maps that Matt showed us in the morning meeting (above.)  The number of hits within a unit will help the archaeology team decide whether to install an archaeological unit (5×5 foot) and take the time and energy to excavate.

I’ll be posting again, time permitting, tomorrow – so check back.