Research and Documentation
Note: I was asked by Robert Reid to build a scale model of a typical soldier's hut as would have been occupied by his ancestor in the winter of 1777-78 at Valley Forge. I had built several boat models for him for which I had some historical knowledge. But I had virtually no expertise in the matter of Valley Forge log huts. In order to guarantee that this model wuld be accurate, I had to undertake a considerable amount of research over a period of months, before and during construction. The following is derived from that research and serves as an explanation, and in some ways a justification, for the way the model was created. Phil Lord 9/12/15
In mid-December of 1777, General Washington's army of over 12 thousand men was in southeastern Pennsylvania making preparations to encamp for the winter. Having been living in canvas tents, the need was to quickly construct hundreds of more substantial shelters out of the raw materials the surrounding forest could provide.
By the time of the Revolution, Americans had already a long history of constructing log structures of a variety of types with a variety of degrees of finish, from direct use of round logs merely cut and dragged from the woods (below, left) , to sophisticated buldings of squared timbers created using the broadaxe and adze (below, right).
Many of these structures could be erected without use of any iron fasteners, such as spikes, nails or screws. But to do so normally required a collection of tools that most farmers, and any professional builder, had at hand, including a two-man cross-cut saw for harvesting logs and trimming to length, an axe for shaping notches and the finer detail trimming, a hatchet for small work, etc. (below).
And critical for making a substantial log structure was an auger for drilling the holes for the wooden pegs and pins that would serve in place of nails (below)
Ideally, a froe and mallet was needed for splitting shakes or shingles out of large sections of log called bolts (below, left), and a shaving horse and drawknife (below, right) for shaping the shingles to final form, which will be discussed further on.
How many of the Army were skilled, or even experienced, in erecting log structures, or the use of such tools, is unknown, but probably there were few, compared with the job they faced of erecting over 1,000 log huts in a few weeks in bad weather. And while the typical log house-building tools could be found on many of the farms in the surrounding locale, when sharing out these tools among over 1,000 hut building projects, they would be considered rare at best.
It can be expected that the average hut building team of about twelve men had axes and shared saws, as such would have been used by the army. But it is also likely that they would have had to beg use of anything more effective. One account records that an entire hut was built with only one axe.
The most important thing to keep in mind when thinking about the building of military winter encampent huts, compared to general late 18th century log structure technology, is that normally log structures were not built in winter, nor were they normlly built in a rush against the clock.
There is very little eyewitness evidence for the building of these huts and only the briefest contemporary documentation, which indicates each hut was to be built by the twelve men intended to occupy it. Details on the intended hut design is contained in Washington's Orderly Book, as follows:
"Fourteen feet by sixteen each; the sides, ends, and roofs made with logs; the roofs made tight with split slabs, or some other way; the sides made tight with clay; a fireplace made of wood and secured with clay on the inside, eighteen inches thick; this fireplace to be in the rear of the hut; the door to be in the end next to the street; the doors to be made of split oak slabs, unless boards can be procured; the side walls to be six feet and a half high."
This general plan leaves open to interpretation many of the details of construction, which is evident in the variety of replica huts built over time. In order to create a diorama that would present the appearance of a hut under construction as it would have been seen in December of 1777, I had to address several of these details, rather than simply copy one of the replica huts seen at places like Valley Forge (above).
There is only one (to my knowledge) eyewitness account of the building of the winter huts at Valley Forge. It derives from the diary and recollections of soldier Joseph Plumb Martin, and was published in the 19th century as "A Narrative of a Revolutionary Soldier" (below, left). The complete text of his description of hut construction is shown below, right, but specific extracts will be inserted where relevant in the following discussion.
The first reaction looking at the model I made is it does not look right. It looks too "new", not "historic" enough.
Somewhere along the way people have developed the attitude that old things...authentic things... need to look weathered. This derives perhaps from the world of antiques, where a distressed and worn finish separates originals from modern reproductions. And when people visit places like Sturbridge Village they expect the boxes and barrels in the back of the general store to be grey, and see any new looking ones as fakes. Yet one must remember that in the year being represented, the boxes and barrels and other items would have been brand new! So recently, living history museums have begun to display their objects as they would have really looked in 1830, or 1778, or 1864, or whatever year is being represented.
Now with structures, and particularly with log cabins, the matter becomes complicated. Surviving originals would be weathered (like seen above, left) because the logs have been subjected to the environment for many deacades or even hundreds of years. So people expect 200+ year old soldier's huts (as shown above, right) should look the same, not all fresh and raw like the model (below).
This perception is inadvertantly reenforced by the fact that replica log huts at historic sites, erected at great labor and expense, are built to the same high standards as a frontiersman would use to provide a home for his family. Durability was essential in both cases, so peeled logs would be preferred over those in bark, and they would be expected to remain in the elements for many years. Unless seen within a few months of their creation, they would appear weathered and grey.
But, in December of 1777 a visitor to Valley Forge would have seen hundreds of brand new huts (as above), completely unweathered. And even the following spring, given the cold weather and snow (not rain), the logs and shingles would still look fresh, since it takes months of warm and wet weather to create the greying one sees on old wood. And never would anyone have seen these huts in a weathered state, as they were dismantled by locals, apparently, as soon as vacated in the spring of 1778.
Most recreated soldier huts, as shown in several places above, have been built out of large diameter peeled logs. While there is some mention of teams of oxen or horses used to drag cut logs out of the forest to the building sites at Valley Forge, three factors suggest smaller diameter, unpeeled logs were probably used in 1777. First, there was massive competition for timber. The need to rush to completion probably meant taking the best logs to be found quickly and transported easily and raised by hand. None of these factors suggests searching for large diameter logs.
Note in the engraving below there are nine logs in the six foot high wall, whereas in the typical replica soldier hut (see pictures above on page) there are only five! That means the logs in this frontier cabin were about 9" in diameter while those in the replica hut are about 14" in diameter!
The typical process in the 18th for erecting a log house in a previously un-inhabited location would be to cut the logs in the spring, peeling the bark easily when sap rising underneath makes it loose, and setting the logs aside while crops are plamted among the stumps on the newly cleared ground. Sometimes sheets of the removed bark were lain on the ground under rocks to flatten them for use later as a provisional roof covering.
Leaving bark on logs to be used in a home is to be avoided as insect infestations can develop in the living bark when exposed to year around moderate temperatures from a heated house. Also, the clay chinking used to seal the gaps between the logs, if resting on bark that later dries and becomes loose, can fall away and cold air leaking in under cracked and loosened bark reduces the efficiency of the walls.
Then in the summer, before the pressures of the harvest, and the demands of the hunt, the structure can be erected with clean and dry timber. And by first snow, the family is housed inside a comfortable house.
But in Valley Forge in 1777, by first snow not a lick of work had been started. December found 12,000 men trying from a cold start (pardon the pun) to replicate the log house building process... completely out of sequence (above).
Harvesting the logs could go forward, but to take the extra time to peel each log, especially when in the fall the bark is tight as the sap is dormant, would be seen as an extreme waste of time. Unlike the frontier settler, these men were urgently building shelters for a few months, not creating homes for a family to last for years. So concerns about insects or cracking and loosening bark, which would perhaps not begin to emerge for years, was not in their minds. Time was of the essense.
There were two popular methods for connecting the logs at the corners without the use of any metal fastenings...just the logs themselves, and an axe.
Of these, the saddle notch would at first seem the simplist, as each log was cut with a concave notch to fit the log above, or below, depending on design. But the end of each log was different, in diameter, so each notch was custom cut, and in "professional" log house construction, a scribing device was used to mark the cut to be made. And trying to shape a custom cut rounded notch with an axe takes skill and is time consuming.
A V-notch, however, has straight sides, so is easier to make with an axe. And since one is shaping both the top of each log and the notch below, they can always be made to match. It is quicker and easier and requires less training or skill.
I used the V-notch system on the model (above, center) for the reasons listed above, extending the logs through and beyond the corners, instead of flush (above, left) because access to saws would have been limited, and use of saws would have added time to the effort for only cosmetic reasons. Having so finely finished a corner (left) makes sense for a family home.....but has no functional purpose in an emergency military operation of limited duration. A detail from a 19th century photograph of an early log cabin (above, right) shows exactly the same corner notching arrangement as on the model.
No matter how deeply and skillfully notched one makes a structure using logs, even if beautifully squared with broadaxe and adze (below, left), gaps occur between the beams, and some form of filling, or "chinking", needs to be applied. This is even more essential when round logs are used (below, right).
Washington's orders state "the sides made tight with clay". And certainly any frontier settler making a future home for his family would be sure to locate and use the best clay he could find. But finding clay deposits at Valley Forge adequate for the caulking of over 1,000 buildings may have been impossible. And collecting and transporting clay in freezing winter conditions would be extremely difficult at best. Such clay as might be available would certainly be reserved for the fireplaces, but even there, options had to be considered. Martin, who was there, observes that the fireplace and chimney should be sealed with clay.... "if clay was to be had; if not, with mud."
The replica huts erected on historic sites present examples of careful and refined application of high grade clay chinking. They are neat, clean and orderly (above).
But soldiers at Valley Forge were cold and tired, had been living in tents, were poorly clothed, and were anxious to get shelter built as quickly as possible. The raw material for chinking was probably mud, created in some pit near the hut site. It is not reasonable to expect these men to take the extra time and care to make attractive seams between the logs, handling a muddy slop in freezing weather with no water-proof gloves, where one's hands would go numb in minutes. Most likely it was thrown onto the gaps between logs by any means, and if shovels were lacking, wooden flats or boards could be quickly shaped to be used in their place (below).
Unlike the frontier home-builder, where appearance and function were wedded in the warm days of late summer, for these men in December of 1777, function....the sealing out of cold air....was the only priority. And given that this make-shift hut, no matter what it looked like, was to replace the "comfort" of a thin canvas tent, aesthetics went out the window (if there had been a window...). I discarded the neat chinking look for a more "sloppy" appearance (below, right), more in keeping with desperate men in brutal cold with freezing hands working gobs of icey mud.
And one can see that even in a homesteader's cabin (above), built no doubt under more favorable conditions, this "sloppy" (but effective) approach was used.
Roof - Framing
Typically, framing a house roof, whether of dimensional timber or logs, involved the same engineering approach. The top of the side walls ended with a capping timber called a "plate", and the rafters were notched to bear onto this plate and extended parallel to the gable end walls to rest against a timber at the peak, often called the "ridgepole" in log structures.
And one can see, in this picture of two of the Valley Forge replica huts, the builders have used that same system of rafters running up from the side walls to the peak, then shingled over. The ends of rafters can clearly be seen (below, circled in yellow).
The only way to fasten runs of shingles to a roof made this way is to have some sort of horizontal stringers attached to the vertical rafters onto which to nail the lines of shingles. And if one looks closely at the undersides of the roofs of these replica huts, the strips of nailing stringers can clearly be seen (below).
The obvious problem with applying this type of roofing to the huts built in 1777 at Valley Forge is that it takes a VAST number of iron nails to attach the many shingles. And by all accounts, nails were as rare as hens teeth at the building site, and such as may have been procured would have been prized for the construction of the doors to the huts, where nails were nearly indispensible.
Here we can turn to the observations of the only known eyewitness to the hut building, Joseph Plumb Martin:
So the "rafters" as such were laid up "parallel to the plates...", meaning they ran the same direction as the logs in the side walls, not at right angles in the examples shown above. Below we see a ruined log cabin, about the same size as these soldier's huts, where the horizontal rafter system is seen, resting on the ledges formed by the shortening of the gable end wall logs.
The reason for the lengthening of the "the two end sticks which held those that served for plates" cited by Martin is evident if we draw out a schematic of the difference with, and without, that extra detail (below).
Looking endwise at the structure, if the plate was just resting on the topmost log in the side wall (left), the shingles would only be supported to a point directly above the wall. But if the end wall "end stick" was extended slightly (right), then the plate or an adjacent log rafter would support an overhang of the outermost shingle, thus providing protection of the wall below from weather.
Another replica hut at Valley Forge shows the correct rafter arrangement (above), constructed on the design observed by Joseph Martin in 1777. The same engineering was applied to the roofing structure in the model (below).
Roof - Covering
Mention has already been made of the manner of making shingles for an 18th century roof, using a froe and mallet to split the shingles from sections of log. These were then trimmed to a taper at one end with a drawknife using a shaving horse (see below) so that they would fit nicely on top of one another to form an even surface.
After starting with a double row at the bottom, shingles, which were then normally three feet long, were overlapped so that the exposed parts of the row above landed over the tapered upper 2/3s of the ones below, so that gaps did not fall together causing a leak (below, as shown for a modern building).
The problem here is this method requires not only some substrate to nail the shingles onto....and the longitudinal rafters Martin describes might be too far apart... but it requires a VAST number of hand wrought, iron nails (above, right). Just this small section (above, left), which is probably only about 36 square feet, takes about 50 nails! One of the soldier's huts at Valley Forge would have a roof area of about 640 square feet. To roof one hut in this manner would take, therefore, over 900 nails. To roof the entire encampment this way would have required well over a million nails!
Lacking nails, the soldiers used an effective alternative method....at least effective enough to get them through the winter. Martin provides the clues as how this was done:
The method of riving the shingles from log "bolts" is nothing different from the technique described above. He does state, however, that the length of the pieces was "about four feet", instead of the standrd three foot shingle of the 18th century. The two replica huts shown before, built at the Valley Forge historical park, relied on the typical 18th century shingle roof technology (below). We saw above they had vertical rafters and horizontal stringers, and here we can see they used standard three foot tapered shingles laid on as single thickness runs, and no doubt used many kegs of iron nails in the process.
But in Martin's account one notices the lack of any mention of tapering those "shingles, or rather staves" using a drawknife and shaving horse. Two reasons are suggested here. First. it is likely drawknives and shaving horses, such as might have been commandeered from the surrounding farms, would not have been sufficient to the demand. But beyond that, spending endless hours trimming piles and piles of shingles, just so they would lay flatter and look nicer, would hardly be seen as an effective emergency measure with hard winter fast approaching. And thirdly, again as said before, these huts were only meant to last a few months, so any method that worked in the short term was plenty good enough.
Another Valley Forge replica hut (above) can be seen to use the system of horizontal rafters, cited by Martin, and they have sheathed the roof with four foot long shingles laid on in double thickeness rows, as he also states (below):
But clearly the shingles in the hut above were nailed onto the rafter logs. And we already have allowed that nailing on the roofing is not correct, or if possible at all in 1777, and would have been reserved for the officers' quarters, not for the huts of the common soldiers.
Martin's use of the words "carefully breaking joints" may seem unfamiliar, or at best an archaic phrase no longer understood. But it remains in use today in the shingling trade, and simply means that the next line of shingles covers the gaps in the previous line so no rain can get through (above).
But if nails were not used, how were the shingles held on the roof?
It is simply the only way to secure rows of shingles without nails, or wooden pegs, which would be very labor intensive and require an auger. This method was commonly used at that time and even later at any time where nails were lacking (see below):
And some later built replica huts, perhaps influenced by Martin's narrative, recently republished, do in fact exhibit this "correct" technique, such as the one below at the Morristown Battlefield site.
This technique was used in the completion of the model (below), with the logs holding the rows of shingles down being bound to the rafters underneath in the manner suggested in Martin's account, when he states "these were then bound on by a straight pole with withes...", which are strips of flexible bark. Given the late season, where bark is brittle, the bark used was probably stripped from very small saplings, or perhaps cut from the inner bark of elm or birch.
But Martin's narrative presents an ambiguity in regard to how each run of shingles related to the "straight pole" which held it down. He states "these (shingles) were then bound on by a straight pole with withes, then another tier with the butts resting on this pole and bound on as before...". If we take this literally, with the bottom ends of each row of shingles resting on top of the log holding down the row below, it would look like the schematic "B" below. But every replica hut built with these sapling "hold-downs" shows the butts resting on the top end of the row of shingles below ("A" below).
A roof built as in "B" above would not be very tight, although it might shed snow and rain well enough, and lacking any other evidence, the model's roof was built to design "A". Perhaps Martin meant by "the butts resting on this pole" as resting against the pole, so that the pole below provided a stop for preventing the shingles above slipping down (like A above).
Any house of substance, log or otherwise, had a fireplace and chimney built of stone (or brick) and good lime mortar. If adequate stone were lacking, as here at Valley Forge, a combination of a stone fireplace and wooden chimney would suffice, with clay often substituting for mortar (below).
And if stone was hardly available at all, the entire fireplace and chimney could be created with wood and clay, or mud, if clay was hard to obtain, and this is confirmed by Martin (below).
This design is seen in the the Morristown replica hut (below, left), the 19th century engraving (below, center), and in the 19th century photo of an old log cabin (below, right). Small diameter logs are used in the replica hut and the engraving, where a barrel serves as the top part of the chimney. The base of the chimney in the engraving (center) is also built of logs and echoes the words in Washington's Orderly Book, which seems to recognize that 1,000 teams of builders would be hard pressed to find suitable stone for fireplaces: "a fireplace made of wood and secured with clay on the inside, eighteen inches thick; this fireplace to be in the rear of the hut;"
But the chimney on the old cabin, (above, right), is not made of small logs but of split logs, forming sticks, and sealed with clay. In doing this research I found chimneys, whether built on a stone base or totally of wood, most often used splits of wood, not small logs. And one reference to soldier huts during the civil war specifically mentions building chimneys of "split sticks". Clay or probably mud, was used to fill the gaps, inside and out. If clay was found it was reserved for the fireplace, where it formed a kind of brick when exposed to the fire, whereas mud might just fracture and break away, risking fire.
The model followed the guidance of the documentary sources and the examples of existing cabins and a fireplace and chimney of split logs and mud (above, right) was created, even though some replica huts were built with wooden fireplace and chimney assemblies made of over-sized logs (above, left)
We have sparse evidence of the interiors of these huts, except to note that twelve men had to be housed inside, plus the brief, and sarcastic, comment left by soldier Martin (below).
We may assume twelve bunks, three high, six to a side, made of poles and split slabs, and then covered with whatever could provide comfort through the night.
The interiors of most of the replica huts reflect this arrangement (above, right) and the model was basically made to duplicate that plan (above, left and below).
The most technically challenging part of the entire log structure was the door. It had to be stable, hold its shape with next to no support, had to be relatively weather-tight, had to swing open and shut, had to be able to be kept in a closed position, yet able to be opened at any time from inside or outside.
The doorway (above) would have had the threshold and header cut into the logs at bottom and at top (above), and this was duplicated in the model. The vertical framing timbers were pegged into the ends of the wall logs to maintain structural stability. If planks were available from a sawmill, as some were in 1777, making the door would be much easier, as planks could be set up parallel, with horizontal battens nailed across the back. (Before, I said if nails were available, the priority would be to save them for the doors.)
But Washington's orders suggest sawn planks might be hard to procure (below):
"...the door to be in the end next to the street; the doors to be made of split oak slabs, unless boards can be procured;"
The model uses split slabs, which would take some time riving out of logs and trimmed to fit. The splits used as battens could be nailed on, or attached using wood pegs, if an auger was available to drill the pin holes.
Iron hinges could not be expected, but a door could be hung with leather straps, again using nails, or even rope. The latch could be carved up of wood if a skilled carpenter were available, or the simple latch and string, often used in early cabins, could be used (below).
In this case a pivoting bar was attached to the back of the door. Anyone with a pocket knife could whittle a bar and latch system (above, left) out of scrap wood. Then a string was attached to the bar and was passed through a hole in the door to the outside (above). This way the door could be opened from either side, and to "lock" the door, occupants merely pulled in the string.
An early American phrase meaning hospitality... "the latch string was always out."
When examining the resulting log hut structure, whether a replica or the model, one sees a lot of locations where it seems cold air would leak in, even with the mud chinking between logs.
Of course the standard for comfort we might expect for a frontier family, with women and children spending winter in a log cabin, is probably vastly superior to that of soldiers who are already hardened to conditions and would consider any sort of structure a great improvement over the tents they had been living in.
But beyond that, we need to consider the mechanics of the "heating system". These huts had an open fireplace, that by accounts would have been kept burning 24 hours a day. Historians have done experiments in replica huts and found that these fireplaces needed a decent supply of incoming fresh air to draw properly, and without it, the huts would fill with smoke. Leaving the door open to aid the fire would seem counter-intuitive. So the moderate leaking in of cold air seems to fit the needs of the fire, as well as preserve the "comfort" of the inhabitants.
This discussion is provided to justify the particular way in which the model Valley Forge soldier hut was made. To view the model in more detail, go to the model page.
Email me at firstname.lastname@example.org if you need more information.