My thoughts on the engine of the Double-armed Man

In my previous post on the subject, I outrageously claimed that I knew how the device worked, if not its exact form. This post will explore the requirements of the device, some other attempts and the thoughts I have on making a reconstruction.

“Meane while the practice of the bow may be received; though we retain the use of the Corslet; for if the one be small incumbrance on the body, and the other will be little burden at the back, and if to rest the left hand on the pike, enable men to draw a stronger bowe, that ingenious device of scruing both together will be best”

– A New Invention for Shooting Fireshafts in Longbowes, Anon, London 1628, p5


There are three primary requirements for a device to connect the bow and pike, and all interpretations should be tested against these requirements:

  1. That is allows full freedom of motion about the horizontal axis.
  2. That it permits no lateral movement in the vertical plane.
  3. That the bow can be “very readily slipped off the Pike, and so trailed therewith”[1]. I’m taking that to mean that it does not present an impediment to the quickest of the drill movements,  so must be repeatably removable within the space of one drum beat.

There are some additional constraints I have applied to the reconstruction. These are obvious, but do help weed out some of the other interpretations:

  1. It must be consistent with the existing pictorial and textual evidence. Yeah, easy one to start.
  2. It must use early 17th century technologies within the ability of village artisans to make or repair[2] and near facsimiles of materials where the original is difficult or impossible to obtain.[3] This is particularly important when the reference to the country Trained Bands is taken into account.
  3. The existence of the patent is evidence that there were thought to be problems with other people copying Neade’s design, or that the technology was reasonably available to do so. The terms of the patent implies that some level of mass-production was happening or was planned. Grinder use is explicitly stated, implying specially designed tooling is acceptable.
  4. Given that Neade was able to equip at least 300 men out of his £600 investment, the device and mounting to the bow and pike must cost less than £2 each in 1620s money. I’m assuming the ones that turn up in the country are not the same 300 that were displayed by the Artillery company, redeployed. This gives me a generous budget of £128.50 (A$190) per unit using the retail price index. Obviously the figure is wrong, but I needed a number to work with.  This buys materials and about two and a half hours of a weaponsmith’s time here and now.
  5. It must be generic, in other words, it must be capable of fitting any bow and any pike conforming to the range of sizes either in statutes or in archaeological finds.
  6. The method of attaching the device to both the pike and bow must be similar to that used to attach parts together in contemporary weapons. For example, Waldman (Fig 79a, p90) shows an X-ray of a mid- to late 17th century halberd with the head attached to the shaft with hammer-driven screws, looking suspiciously like modern countersunk wood screws.

Previous interpretations

Others have attempted this task before me, with varying degrees of success. Most are archers, and thus lack experience in the finer points of seventeenth century military pike drill. Generally, their interpretations are only created for a modern target bow of 35-50 pound draw weight and neglect the 60-70-pound Stuart  hunting bow or the heavier draw war bow, both of which are adequately documented in use by individual members of the London Trained Bands.

Neade: The Pikeman stands ordered.
Note the small shaft connecting the bow and pike near the top of the woodcut.

Many of my predecessors came up with a device that consisted of a ball-joint or universal joint attached to two sleeves, one slid on to the pike, and the other to the bow. There are a couple of problems with this approach: the sleeve must be custom built for each bow and each pike, ruling out the type of mass production hinted at in the patent (and in Neade’s entreaty to the king about bankrupting himself setting up a factory) and; the bow can not be removed from the pike, causing problems with the drill postures, such as when the pike is palmed or trailed when the bow is supposed to be “very readily slipped off the pike… and with as great readiness slipped on againe” [4]. Any non-rotational movement in the joint will allow the pike to flop, and rather than serve “as a Rest for the bow arme, whereby we draw a very strong bow”,[4]  and add to the load on the bow hand, constantly twisting and pulling the bow off line.

Illustration from Ward’s Animadversions of Warre (1639)
The top pike is fitted with musket barrels, the lower with a bow.

The “sleeved bow and pike” approach, while fitting Neade’s woodcuts at a stretch, ignores Ward’s depiction showing a circular mounting plate on the side of the pike, and makes no allowance for the natural variation in size of the riser on hand-made bows, such as would have been the norm in Neade’s time: Markham and Ascham both state that the bow has to be custom made to suit the available timber and the archer. Hugh Soar addresses this by allowing the sleeves to be varied in diameter.[5]

The other approach has been to rigidly attach the bow to the pike (or often spears) by means of metal collars or in some cases, simply tying with a square lashing. I briefly exerimented with a threaded end to the axle off the bow and literally screwing it into a threaded nut mounted on the pike, but couldn’t come up with a system that was secure and could be separated in the time of a single drum beat. My best effort was three drum beats which became a major encumberance when the command to trail the pike is given. Spinning the bow around to unscrew it requires a 3′ radius space, which doesn’t exist in a pike block on the march and a delay of three beats while you spin your bow ends up with your pike tangled in the trees or powerlines.

The “rigid attachment” school, while successful for archery, particularly with the shorter spears[6] and recurve bows often used by these people, comes unstuck once the pike is moved to any of the horizontal or vertical postures, such as the charge, advance or order. Shouldering or trailing the pike is impossible with this configuration. Neade stipulated that there should be no encumbrance to the pike postures.

So let’s review the requirements and see what we can come up with. The numbers match in the the lists above and below.

  1. To be consistent with the pictorial evidence, we need to join the the bow and pike with a short bar or axle that allows one to rotate with respect to the other, but doesn’t allow any movement in the other two planes. It looks like it isn’t more than an inch or two.
  2. This part of the 17th century was just on the cusp of the early part of the industrial revolution. Much of the metallurgy and energy technology necessary for this to happen was understood, but manufacturing was still small scale. Cast iron was becoming more common, but would have been impossible to repair anywhere other than specialised foundries, so our materials are likely to be cupric or wrought iron or steel, easily worked by a village smith. It would be more easily repairable if it used parts that were already produced for something else.
  3. Fear of copying implies the design is simple, or if complex, can be made relatively easily or cheaply using fairly common tooling. The patent covers both the device and a grinder so a particular size, shape or guide so some tooling may be necessary. Hand-turned or power driven grinders were relatively common and a feature of most workshops and farms, so the use of a grinder in itself is not remarkable.
  4. My cost calculations of A$190 per unit buys materials and about two and a half hours of a weaponsmith’s time here and now. While not directly comparable as labour cost is now higher than material cost rather than the other way around, it does give some idea of time and complexity. I’m going to go out on a limb and suggest under a military influenced economy parts from the standard Tower-pattern muskets  would be a cheap and easily available. I’ve priced these parts from modern gunsmiths and think that after buying all necessary parts, I’d still be able to get 90 minutes from the smith.
  5. and 6. The generic mounting requirement means the method of attachment must not depend on an exact match with the shape and/or size of either the bow or pike. If we take the cue from Ward, it appears to be some sort of disk shaped mounting plate, possibly attached by screws. Neade’s woodcut shows binding, which may obscure the same method as Ward, or may be a different system that allows adjustments. In conversation with a couple of the armourers at the Royal Armouries, the use of pikes in decorative wall displays in historic houses and the various screws, bolts, various other attachments and coats of paint have destroyed any chance of being able to identify a pike that once had a bow attached.

At this point, we’ll leave history behind and dive into possible pasts instead. As much as I’d like to pretend that this will be Neade’s device, it will really be mine of similar function. It will be as close as I can get, like Neade, I use a longbow, practice pike drill with copies of original weapons and in copies of original armour (copied from a suit made by a possible ancestor of my wife, but that’s another story)  and can apply my engineering training and understanding of the properties of the materials involved to the design. Of course, the intervening 400 years have created more differences than commonalities but we’ll just have to live with that.

My design consists of two round flat mild steel disks, one to attach to the pike and one for the bow. A solid axle of about half an inch in diameter and 1.5 to 2 inches in length projects from the centre of the disk attached to the bow. A square groove is cut right around the axle about 0.25 inch from end nearer the pike. The locking mechanism will fit into this groove. A sleeve of similar length and wide enough to slip over the axle is attached to the disk that is mounted on the pike. This sleeve has a window cut in it, going for a quarter of the circumference of the sleeve and the width of the groove in the axle.  The locking mechanism is attached to the pike plate and consists of a scear and spring from a Tower-pattern musket. The scear is a simple cranked lever and fits through the window in the sleeve into the groove in the axle. The spring provides positive pressure to keep the scear in place, much like in the musket the spring acts on the scear to keep the trigger and match in the not-going-bang-position.[7] The assembly is attached to the pike with screws or nails, and to the bow by binding because I’m paranoid about drilling holes in bows and my bow is somewhat smaller than a Stuart warbow and possibly a damn side harder to replace.

This design looks superficially like both Neade and Ward’s pics, uses common 17th century metalworking techniques, tools and materials, and commonly available matchlock parts that can be serviced by just about anyone. Manufacture in 90 minutes isn’t out of the question for even the most basic smith, my first build using hand tools and a power-grinder to put the groove in the axle and window in the sleeve took about 120 minutes but I’m not a particularly accomplished smith and was refining the design as I went. It also meets the requirements for fitting all the pikes and all the longbows that I have access to by simple expedient of bending the round disks a bit. In modern WHS parlance, there is a pinch point where the open end of the sleeve meets the bow, but the danger can be minimised by making the gap as small as possible and could be further helped by embossing the disk to raise the hand slightly off the sleeve end.


[1] Objections

[2] Held, p51

[3] … so no potatoes. Mild steel is used as an alternate to wrought iron, even though the properties aren’t a close match

[4] Objections

[5] Soar, Crooked Stick, p112

[6] ie, shorter than the statutory 17 feet

[7] but not that not-going-bang position known as a hangfire


Atkin, M., Worcestershire Under Arms – An English County During the Civil Wars, Pen & Sword Books Ltd, Barnsley, 2004

Anon., Objections to the use of the Bow with the Pike: And the Answers thereunto, London, 1636(?)

Anon., A New Invention for Shooting Fireshafts in Longbowes, London 1628

Bariffe, W., Military Discipline or The Young Artilleryman, London, 1643

Charles R., Patent AD 1635, No 69. Affixing the Bow and Pike Together, London, 1635

Charles R., 9.Car.1 A Proclamation for the use of the Bowe and Pike together in Military Discipline, London, 1633

Charles R., 10.Car.1 De Concessione Privilegii Willielmo Neadee, London, 1634

Credland, A. G., Bow and Pike, Journal of Archer-Antiquaries Vol 29-1986

Neade, W., The Double-armed Man, By the New Invention, London, 1625

Roberts, K., A New Combination – English Experiments in Infantry Equipment, 1620-40, Military Illustrated 90

Soar, H. D. H., The Crooked Stick, Westholme Publishing, Yardley, 2004

Stone, B., Derbyshire in the Civil War, Scarthin Books, 1992

Venn, T., Military and Maritine Discipline, London, 1672

Ward, R., Animadversions of Warre, or a Militaire Magazine of the Truest Rules and Ablest Instruction For the Managing of Warre, London 1639

Waldman, J., Hafted Weapons in Medieval and Renaissance Europe – The Evolution of European Staff Weapons between 1200 and 1650, Brill, Boston, 2005.


2 thoughts on “My thoughts on the engine of the Double-armed Man

  1. Hi Tony, I haven’t got that far yet due to not being able to source either a reliably accurate spring, nor a blacksmith that could understand what it was that I was carrying on about. It’s one of these fullness of time projects.
    I had made a wooden model, but as it didn’t have a full length bow or pike attached, nobody I showed understood what they were looking at.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.