The following article was recently published in ‘American Period Furniture, 2013′, the annual publication of the Society of American Period Furniture Makers . I wrote the article to chronicle my study of Joseph Smith’s ‘Key’. I also thought I would post it here. If you are not a member of SAPFM, you should join! Enjoy…
Joseph Smith was an English engraver working around the turn of the 18th century, and he is noted as having produced one of the Western world’s first tool catalogs, called formally, ‘Explanation or key, to the various manufactories of Sheffield, with engravings of each article’. He was not a tool maker, or dealer, as far as we can tell (Kebabian, 1975), and he was not a cabinet maker, joiner, or even a woodworker. Quite ironic then, that his work stands as perhaps the one of the most important historical records of the tools of period woodworkers.
The ‘Key’, first published in 1816, contains hundreds of images of all forms of edge tools from chisels and axes to planes and boring bits, not to mention flatware for the ladies of the time. The plate featuring the handsaws is quite profound, for it not only presents the images of the tools, but it names them and gives the available dimensions of each saw at the toothline. And unlike the saws featured in Moxon, Fellibien, or Nicholson, the detail is outstanding.
I cannot recall when I first beheld the handsaws featured in Smith’s ‘Key’, but I do recall quite vividly how I felt. I was mesmerized. I had never seen anything like them before, and I would spend the next several years trying to understand them in every way that I could. But like any piece of archeological evidence, while the plate answers some questions about what period saws looked like, it asks so many more, such as how accurate are the engravings to begin with? Are the details like the tapering of the backsaw plates to scale? I also wonder if Smith used a particular saw maker’s wares as models for his engravings. This question may well be impossible to answer. And what of the reality of period engravings in general? One need only view Smith’s plate featuring the fillister and plough planes to have serious doubts about his ability to accurately capture a tool in proper dimension and scale. Are the saws in Smith’s ‘Key’ accurate?
After gazing at that single plate for hours on end, I wanted to find out these answers. And I wanted to know not just what the saws looked like, but I wanted to know how they felt in the hand, how they cut, and how they were made. So I set out to find actual saws matching the forms shown in the ‘Key’. Pictures of period saws would not do…I needed to hold the saws in my hand. I wanted to feel the grain and patina of the tote…smell the tarnish on the brass back…and know the weight and balance of the saws in my hand.
In my searches, I set out to find saws made in Sheffield between 1800 to 1820. I use this twenty year span based on three factors: 1) It is unclear as to when Smith actually engraved the plate featuring handsaws. It could have been as early as 1801, when he was commissioned by Peter Stubbs, noted Sheffield tool maker and wholesaler to create several plates for a catalog, but it could have been as late as 1816, when he himself published the entire ‘Key’ (Kebabian, 1975). 2) It is difficult at best to date any saw from the period to within a few years of its actual date of manufacturer due to the extremely limited amount of documentation and business records available. And finally, 3) Since tool forms were based on an entirely tradition bound culture, evolution was relatively slow. A common saw made in 1800 would likely feature the same shape and details as one made in 1810 or even later. It wasn’t until the early 20th century that we see the rapid changing and innovation in tool forms to respond to a more competitive market. And so, based on those parameters, over the last four years I found several examples of saws that match the forms shown in the ‘Key’.
I began with the 26 inch handsaw featured by Smith. I have collected half a dozen 26 inch handsaws over the last four years likely made in Sheffield and in the target years of 1800 to 1820. I have also handled twice as many more belonging to other collectors. This relatively large sample offers a rich palette of information for my study. The features of the Smith’s handsaw that distinguish it are as follows:
1) London pattern tote featuring a flat bottom hand grip.
The London pattern was a popular aesthetic detail of the time for saw totes. It was standard fare for most saws made in the late 18th and early 19th century. It is distinguished by the flat at the bottom of the tote grip, as opposed to the fish tail pattern that became the standard by the mid-19th century. Beech is almost always the wood species of these totes as well.
2) A protruding, narrow boss atop the tote cheek.
I use the term ‘boss’ to describe the peninsula of wood that protrudes from the foremost upper section of the tote. It is quite narrow for saws of this time, and is easily distinguished from later saws which feature a much wider protrusion, and which curve back down towards the toothline. Saws earlier than about 1780 or so do not feature this boss. Though saws earlier than this are quite rare, those known saws feature a completely rounded cheek and typically no protrusion at all.
3) A straight back saw plate with nib and soft radius on the nose.
The nib was also standard on almost all handsaws prior to the late 19th century, so it is of little help in dating an early saw. However, the shape of the nose of the saw plate is quite distinguishing. One can almost see the evolution from a very broad rounding at the nose (radius of 2 to 3 inches) to an almost square edge (radius of ¼ inch). The Smith’s saws have what could be determined to be a mid-point in the evolution from round to square. It appears to be a radius in the neighborhood of 1 inch.
Viewing the six full-size handsaws I was able to find as a whole, I was struck by a singular thought: they are remarkably uniform. Though each has slight variations in hang, detail, and execution of certain elements, it almost appears as if each maker was working from a singular and common pattern. This uniformity is certainly not unique to handsaws of this time, as we see this in tools more and more as we progress into the Industrial Revolution. Indeed, as many scholars have said before, the western world was becoming standardized. It seems handsaws were no exception.
So what does the collective uniformity of these saws, and their striking resemblance to the Smith’s plate, mean? To my mind it is a validation of the saw engravings on a whole. If Smith was accurate enough in his work to represent the handsaw so well, as clearly my found examples suggest, then it stands to reason that he represented each of the other saws just as well. And therein lies the value of Smith’s ‘Key’: these are not artistic interpretations of tools, or period representations limited by the engravers stroke and the printers plate, but near photographs of actual tools from that time. Knowing this, I was encouraged all the more on my quest to find examples of the other saws featured on the plate.
Not surprisingly, however, finding examples of the remaining forms proved much more difficult. Any saw collector will tell you that they find ten handsaws for every one backsaw, and as well, nine of those handsaws are 26 inches or more for every one panel saw (a handsaw of 16 to 24 inches at the toothline). I did most recently find this little panel saw at an auction and nearly fell over myself in excitement before acquiring it.
The tote bears an uncanny resemblance to the panel and grafting saws featured by Smith. This example can easily be dated to circa 1800, and may even be earlier. This tote is a great example of what I call the ‘swept-forward’ design. The whole shape seems to lean forward slightly…just as in the ‘Key’. It is a subtle element, but one that is crucial to dating the tool.
The saw blade of this small saw is not so encouraging though. It may well be a replacement, but I’m not convinced one way or the other. The marks on the plate suggest a very primitive working of the steel. It was likely rolled, as steel plate had been for a very long time by the 18th century, but it is crude by our standards. There is a small chance it could have been forged by a blacksmith on an anvil, but these saws are very rare even by 18th century standards. The later steel nut and bolt is obviously a replacement on this saw, but this was a very common repair to saws of this time. The hardware of the late 18th century was nowhere near as robust as that used in the 20th century and often broke or wore out soon after manufacture. Regardless, the saw is the real gem. The value of it lies in the tote itself….it is another validation of the form shown in the ‘Key’.
But what of the backsaws? Interestingly, this is where things get blurry. I have not been able to find an open-handled backsaw matching either of the forms in the ‘Key’. This includes both the dovetail saw and the carcase saw. This is not surprising as small backsaws of this vintage are exceptionally rare. They are delicate tools and do not survive easily. I have found two saws of roughly the same time period, but one is a bit earlier and one a bit later.
This dovetail saw, with its completely rounded tote cheek can be dated to the later 18th century, perhaps even as early as 1770. The Smith’s dovetail saw features a pointed cheek which became en vogue around the turn of the century. And the hang is far different.
The other dovetail saw I have found can be dated to the early 1820s or so, but the hang is much more horizontal. Hang is a term used to describe the angle of the tote relative to the back. It can generally be anywhere from 0 to 45 degrees. The earliest handsaws had a near perpendicular hang (0 degrees from the back). As time passed through the 18th, 19th and into the 20th century, the hang of most saws rotated more and more horizontal until it settled at about 35 degrees on most backsaws. It is the hang of Smith’s backsaws that I was so….hung up on. That, and the extreme cant, or tapering in width, of the saw plate along its length are what makes these saws so unique in form.
Of the larger backsaws–the sash and tenon saw–I have had similar mixed results. I have found several 14 inch backsaws of near identical form, but not one as early as the ‘Key’. They are more mid-18th century. I’ve also fared the same with the tenon saw. I have three examples, one of which is a good match, but all date from 20 to 30 years after Smith’s work. I have found one 18 inch backsaw made by Kenyon Sykes & Co of Sheffield that can be dated to c.1815. The saw is a fabulous example of a near extinct form, but alas, it features the rounded cheek common to earlier saws, not the pointed upper cheek of Smith’s.
This 16 inch tenon saw is strikingly similar to the one in Smith’s, but the maker’s mark indicates it was sold by Thomas Tillotson of Sheffield, who was in business in the 1850s. I think we can safely assume that the form was valid and endured some decades after Smith’s work.
But what of the carcase saw? There are examples here and there of contemporary saws…Benjamin Seaton’s chest comes to mind, as well as some of the saws in the Williamsburg collection, but they are not exactly the same. I was hell bent on trying to find one of these examples (without having to re-mortgage my house, of course), but they have proven elusive and just too rare. Left without an actual example to hold and examine, or even a picture to reference, I had only one option: if I wanted a carcase saw like Smith’s, then I would have to make it myself.
At this point, I felt reasonably confident to use the ‘Key’ as an accurate and scaled image of the actual tool. But before I could begin to shape the wood and metal parts, I would have to have a pattern to follow. And this was the truly exciting part…taking the image from the ‘Key’ and translating it into scaled patterns for the distinct parts. Those distinct parts are the tote, the back, and the saw plate.
I began with the tote, as this is the heart of every saw. To start, I enlarged the image from the ‘Key’ on a photocopier until the tote grip measured 3 and ½ inches between the horns. This is a roughly standard measurement on backsaws. My intent was not to simply copy the tote, but re-draft it with a bit of creative license to ensure its functionality and balance. You can see where I adjusted certain elements…
From here it was a simple process of adhering the pattern to an appropriate piece of quarter-sawn Beech and roughing it out on a bandsaw. I used stock of just under an inch thick, as this is a fairly standard dimension.
Next, I turned my attention to the back. The first couple of years that I stared at the ‘Key’ I was so focused on the taper of the saw blade, that I never even noticed the taper on the back. You will see that the back appears to taper from its tallest at the tote down towards the toe, where it is roughly 10 or 20 percent shorter. This would turn out to be a major element in the wonderful function of the saw, and I’m glad a fellow saw maker pointed it out to me. I decided to use a ¾ inch tall brass back and tapered it down to roughly 5/8ths of an inch at the toe. This was done with a hacksaw and files….a very laborious process!
Finally, I laid out the pattern for the saw plate, or blade, as it is also called. Again, just as with the back, it was crucial to get the proportions of the tapering in height, called ‘cant’, just right. I tapered the height of the blade from roughly 3 inches at the heel down to 2 and ½ inches at the toe.
After completing construction of the saw using traditional brass split saw nuts and polishing all of the parts, I turned my attention to the teeth. I decided to file the saw with a hybrid tooth pattern having 10 degrees of rake and 10 degrees of fleam. This allows the teeth to cut both with the grain for ripping and across the grain for cross-cutting. The teeth are spaced 14 points to the inch.
It took about two months of drafting, roughing out, shaping, refining and filing to create my first prototype of the carcase saw. Finishing it was a spectacular moment…anticipation, excitement and pure joy when I ran it through wood for the first time. Not only was I not disappointed with the results, I was blown away. The saw functioned with a refinement and precision that I could have never imagined. The tapering of the saw plate, coupled with the tapering of the brass back serve to lighten the saw as it progresses towards the toe. This lightening puts the mass of the tool more towards the heel and the tote. Much like a mid- or rear-engine sports car, this shifts the mass back towards your hand in the tote, so that in use it feels more connected to your hand and your arm. The result is that the saw feels like an extension of yourself. As such, it is more responsive, requires less conscious effort to push, and it much easier to track accurately.
Besides this taper, the cant of the saw tote is also unique. Most later and now more popular saws feature a hang more in the neighborhood of 35 degrees. This saw features a hang of about 20 degrees. This puts the thrust more behind the teeth in stead of above them. So instead of pushing them down into the work (which mostly frustrates the sawyer and disrupts accuracy) it pushes them forward and lets the weight of the saw create the necessary downward force to cut. This makes the saw smoother, more naturally accurate, and easier to use in general.
Without a doubt, the results of this experiment have been astounding. I have discovered a form of tool that I find superior in every way to the more common saws of later American makers. The only conundrum I find remaining is which saw from the ‘Key’ should I build next?
-Matt Cianci, The SawWright