The Evolution of Useful Things: How Everyday Artifacts-From Forks and Pins to Paper Clips and Zippers-Came to Be as They Are.

by Henry Petroski

  A successful paper clip begins with steel wire that wants to spring back to its original shape after being bent, but only up to a point, for otherwise the paper clip could not be formed into the clever and pleasant-looking object that it is. Steel and all materials are said to behave “elastically” when they stretch, bend, or twist in proportion to the force applied to them and resume their original shape after being let go, as observed by the English physicist and inventor Robert Hooke, who discovered in 1660 the principle that now bears his name. He did not publish it until 1678, however. Even then, in the manner of his times, which included fierce competitiveness over claims to priority, Hooke did not actually articulate the principle but merely published it in the form of the Latin anagram ceiiinosssttuu. Two years later, when he was so inclined, he rearranged the letters into the phrase Ut tensio sic uis and explained that “as the tension so the power” of a spring meant that the more you pull the more it resists—until you pull too hard and the spring gives and does not completely return to its original shape.

  Forming a paper clip presents a common dilemma encountered by engineers and inventors: the very properties of the material that make it possible to be shaped into a useful object also limit its use. If one were to try to make a paper clip out of wire that stayed bent too easily, it would have little spring and not hold papers very tightly. On the other hand, if one were to use wire that did not stay bent, then the clip could not even be formed. Thus, understanding the fundamental behavior of materials and how to employ them to advantage is often a principal reason that something as seemingly simple as a paper clip cannot be developed sooner than it is.

  Steel wire was still new in the second half of the nineteenth century, and early wire manufacturers looked for applications of their product. Some, like John Roebling, went so far as to promote, design, and build suspension bridges, which used large quantities of wire in their cables. (The elastic springiness of large bridges is often very noticeable to motorists stopped in traffic upon them. If, in the process of erection or use, the steel cables were stretched beyond the limits of Hooke’s Law, the bridge would sag permanently like a melted plastic model of itself.) But, whether spinning bridge cables or bending wire into fasteners, specialized machinery was essential for exploiting the new material. To have formed paper clips one by one by hand would have made them very expensive and hardly a challenge in business applications to the modest machine-made straight pin. Hence the widespread manufacture and use of the paper clip had to await not only the availability of the right wire but also the existence of machinery capable of tirelessly and reliably bending it in a flash into things that could be bought for pennies a box. In the meantime, though there may not have been an outcry of complaints about desk pins, there no doubt were numerous inventors and would-be inventors who found the pin an unsightly and inappropriate paper fastener and thought there must be a better way.

  As with many new devices, especially ones of modest proportions and few pretensions, the origins of the first bent-wire paper clip are not without their uncertainties, including those induced by chauvinism. According to an oft-repeated account, a Norwegian named Johan Vaaler should be credited with the invention of the paper clip in 1899. However, as the story goes, Norway had no patent law at the time, and though Vaaler’s drawing was accepted by a special government commission, he had to seek an actual patent in Germany. Norwegians are said to have remembered proudly the humble item’s origins in their country when, during World War II, they “fastened paper clips to their jacket lapels to show patriotism and irritate the Germans.” Wearing a paper clip could result in arrest, but the function of the device, “to bind together,” took on the fiercely symbolic meaning of “people joining against the forces of occupation.”

  Vaaler’s fin-de-siècle notion was granted an American patent in 1901, and the document describes the “paper clip or holder”:

  It consists of forming same of a spring material, such as a piece of wire, that is bent to a rectangular, triangular, or otherwise shaped hoop, the end parts of which wire piece form members or tongues lying side by side in contrary directions.

  As if to emphasize that a paper clip need take no unique shape, several styles are illustrated in Vaaler’s patent. (Such a multitude of ways to achieve the same end is common in patent applications, which thus provide counterexample after counterexample to the claim that form follows function.) Even if Vaaler’s paper clips look superficially like today’s, they differ in one major respect: the wire does not form a loop within a loop. Papers would be held together by the arms of the clips, of course, but they would have required a very deliberate action to apply. Curiously for such simple devices, neither Vaaler’s nor most other contemporary patent applications for paper clips included a model.

  Johan Vaaler’s first American patent, dated June 4, 1901, showed several embodiments of a “paper clip or holder.” The version labeled “Fig. 12” suggests the beginnings of what has come to be known as the Gem paper clip, but is clearly not a fully formed Gem. (photo credit 4.3)

  Vaaler did make explicit a secondary feature of his invention: “To obviate the clips hanging together when being packed up in boxes or the like, the end of one of the tongues … may … lie close up to the base part of the other tongue.” In other words, there were no great protuberances. Anticipating such an inconvenience as clips’ “hanging together” would show uncommon prescience on the part of an inventor, and Vaaler’s mere mention of this problem suggests that other paper clips not only were already in existence but were annoying users with an undesirable feature not unheard of today.

  There were indeed other paper clips by the time Vaaler’s American patent was issued, and it appears to have been granted more for his variations on some common themes than for any seminal contribution. Matthew Schooley, a Pennsylvanian, filed a patent application in 1896 for a “paper clip or holder which, while simple in its construction, is easy of application and certain in the performance of its functions.” Evidently, even at that time the shortcomings of such a device were known, for according to the patent, issued in 1898:

  I am aware that prior to my invention paper-clips have been made somewhat similar to mine in their general idea; but so far as I am informed none are free from objectionable projections which stand out from the papers which they hold.

  Furthermore, unlike a Vaaler-like design, Schooley’s clip would lie “fiat upon or against the papers which it binds together [without] puckering or bending” them. It achieved this by overlapping the wire in coil fashion. Though there is no turn within a turn, its form suggests today’s paper clip almost as much as Vaaler’s does.

  When all is said and done, any attempt to sort out the origins and the patent history of the paper clip may be an exercise in futility. For there appear to have been countless variations on the device, a great multiplicity of forms, and some of the earliest and most interesting versions seem not to have been patented at all, which is perhaps not so surprising for such a modest artifact. Nevertheless, however obscure their provenance, there is little doubt that alternate forms of the artifact evolved in response to the failure of existing forms to reach perfection, and therein lies the value of this most common object as a case study of how failure can drive form to fanciful extremes in quest of parallel objectives.

  In 1900 an American patent was issued to Cornelius Brosnan of Springfield, Massachusetts, for a “paper-clip,” which has been regarded in the industry as the “first successful bent wire paper clip.” Again, no model was submitted, but two versions of the clip were shown in the patent drawings, and they are evocative of the outlines of track layouts that graced the Lionel and American Flyer model-train catalogues that I pored over as a child. Nevertheless, in the typical manner of patent literature, Brosnan’s description of “certain new and useful Improvements in Paper-Clips” suggested the problems with existing ways of fastening papers that the new clip would overcome:

  This invention relates to an improved clip or binder
for fastening together sheets of paper, the object being to provide a form of paper fastener which may be rapidly and cheaply made in large quantities from spring-wire, having the capability of being very conveniently applied in its fastening engagement with several sheets of paper, of holding the papers together with all required security, and yet of permitting its disengagement when desired.

  An American patent for a bent-wire “paper clip or holder” was issued in 1898 to Matthew Schooley, thus predating the commonly cited “invention” of the paper clip in 1899 by the Norwegian Johan Vaaler. Just as Schooley’s patent drawings show different embodiments of the clip, so there are believed to have existed many other (unpatented) variations, some dating from as early as the 1870s. (photo credit 4.4)

  In 1900, Cornelius Brosnan was issued a patent for a paper clip that removed one of the principal objections to many earlier designs. Because the inner leg of this Konaclip terminated in a tight loop, or “eye,” it did not have a sharp end to catch, scratch, or tear the papers it held. However, Brosnan implicitly admitted a failing of his own Konaclip by taking out another patent five years later for a clip that did not have an eye to hook in the loop of a box mate. (photo credit 4.5)

  Clearly, at least in the minds of the inventor Brosnan and the patent examiner, the new paper clip was superior to existing devices, and its unique form was described in three separate claims, each of which began:

  A clip or paper-fastener constructed of a single length of wire bent to form an elongated frame with an end portion of the wire deflected inwardly within and near one end of the frame, and extended longitudinally along and within the middle of the device …

  The claims went on to specify that the wire was “formed corrugated, and terminating in an eye … near the other end of the frame.” This eye kept the clip from scratching or tearing the papers it was attached to, something Schooley’s and Vaaler’s clips were always liable to do. Brosnan did have something: his clip, which was called the Konaclip, took full advantage of the latest technology to bend wire into tight loops, and far surpassed anything then patented. It was certainly easier to apply than most other existing designs. Even so, the Konaclip did not last long, for, notwithstanding his promise that papers would not slip from it, they did, especially the middle ones in a pile.

  Brosnan, like many other inventors, no doubt thought he had spelled out his claims in such a way as to cover all practical methods of bending a piece of wire to serve as a “perfected” paper clip. But perhaps nothing mocks the cliché that “form follows function” so emphatically as this common object. The eye formed at one end of the Konaclip seemed essential, for example, because if the inside of the clip had terminated instead in a straight piece of wire, this could have snagged and pierced papers as the clip was attached, thus reducing any advantage over the pin. But in spite of Brosnan’s advertising claim that his clip, as the final result of a long process of improvement, provided “the only satisfactory attachment of papers,” and his warning to businesspeople, “Don’t mutilate your papers with pins or fasteners,” the fact remained that papers still slipped out from paper clips. Besides, Brosnan’s designs were sure to snag one another in the box.

  In 1905 Brosnan was issued a new patent for a paper clip “of novel shape” that was “formed of spring-wire and constructed with portions which [when] distended oppositely from each other develop a reaction to embrace and bind marginally sheets of paper.” This clip did not depend upon overlapping wire for its gripping power, for, like the Konaclip, it was formed all in a single plane. Rather, the papers were grasped by the spring action created by separating the touching inner and outer loops of wire. According to Brosnan’s patent, his new clip had the advantages of “cheapness of construction, ease of manipulation …, efficiency in the retention and binding action … without liability of swinging or shifting from its given set position, and … not becom[ing] interlocked one with another to cause bother and delay in taking one or more from the box …, [and not catching] other papers in conjunction with which the clipped bunch of papers may be brought.” It is clear what the disadvantages and failings of existing paper clips were.

  Several of the many alternative forms to the paper clips that Brosnan and other ingenious wire benders came up with are recorded in the pages of Webster’s New International Dictionary. As if to emphasize the importance of the form of a paper clip and the difficulty of defining it in words alone, the definition is illustrated. The first edition, dating from 1909, defined “clip” as “a clasp or holder for letters, bills, clippings, etc.,” and showed a predecessor clamplike device along with Brosnan’s Konaclip and some alternative ways to bend wire that he had not anticipated in his patent claims. These clips, which have come to be known as the Niagara and Rinklip, demonstrated, for example, that one did not need an eye or a loop terminating within a wire frame to make the thing function. When the second edition of Webster’s was published, in 1934, “paper clip” was defined as “a device consisting of a length of wire bent into flat loops that can be separated by a slight pressure to clasp several sheets of paper together.” The reader was referred to the “clip” entry for an illustration that no longer included a stamped-metal style or Brosnan’s Konaclip, but showed still another way of forming a clip—with two “eyes” outside the main body of the clip. This design was to evolve into one with its eyes inside the wire frame, where they were less likely to snag their box mates, a design that would come to be known as the Owl style. In early advertisements its superiority to the one-eyed Konaclip was proclaimed in verse:

  The illustration accompanying the definition of “clip” in the first edition of Webster’s New International Dictionary (1909, top) included a device stamped out of sheet metal and a Konaclip. In the second edition (1934, bottom), the sheet-metal clip was replaced by an early version of an Owl-style clip and the Konaclip was replaced by a Gem. The paper clamp, as well as the Niagara (extreme right, top) and Rinklip (extreme right, bottom), both of which are easily applied to papers, retained their specialized positions among a growing variety of paper clips. (photo credit 4.6)

  An eye for business

  One too few

  Observe this clip

  This clip has two.

  The advantages claimed for the Owl clip included—besides that of not tending to get tangled with others of its kind—the absence of any sharp ends that might snatch at papers which did not belong with the clipped pack, or might rip papers upon removal. As matters turned out, however, it was not the Owl that was destined to drive out the Konaclip.

  One of the other clips illustrated in the second edition of Webster’s became known as the Gem, and has now long been the most popular style. Indeed, to the vast majority of people today it is virtually synonymous with paper clip.

  The Gem paper clip did not just quietly develop between the first and second editions of Webster’s Unabridged, however. Indeed, in spite of the Norwegian claim to have produced the archetypal paper clip, the idea of the Gem was fully formed by the time of Vaaler’s patent. It existed, at least on paper, as early as April 27, 1899, for it was on that date that William Middlebrook of Waterbury, Connecticut—a center of mechanized pin-making—filed a patent application for a “machine for making paper clips” and showed a perfectly proportioned Gem as the product of the machine. Since Middlebrook patented only the machine and not the clip itself, the Gem design may have predated his application and been already known to those practiced in the art.

  Even though the paper clip shown in Vaaler’s 1901 U.S. patent was not nearly so fully developed as the one pictured with Middlebrook’s machine, American manufacturers have remained cautious about making direct claims concerning the role of their predecessors in inventing the Gem. An anonymous history that appeared in a 1975 issue of Office Products does describe Brosnan’s 1900 patent Konaclip as a “direct ancestor of the Gem pattern,” but Middlebrook’s 1899 patent shows clearly that the ancestry was at best reversed. Another history—published in 1973 by a Smi
thsonian Institution staff member who would not admit to being a curator of paper clips but only to being “merely their protector”—states that no patented clip was “overwhelmingly successful until the beginnings of the 20th century when the Gem pattern paper clip was introduced.” The language is understandably ambiguous and vague as to the Gem’s patent status and nationality, and this case study points out clearly the limitations of relying entirely upon the patent literature for tracing the evolution of artifacts. A search through U.S. patents for paper clips alone will never turn up an unadulterated Gem, and Middlebrook’s patent for a machine for making them might easily be passed over as not directly relevant to the form of the artifact being manufactured.

  Although the 1899 patent issued to William Middlebrook of Waterbury, Connecticut, was for a machine for making wire paper clips rather than for the clip design itself, Middlebrook’s drawings showed clearly (especially in his Fig. 8) that what came to be known as a Gem was being formed. This style of paper clip, which seems never to have been explicitly patented, came to be the standard to be improved upon. While functionally as deficient as myriad other styles, its aesthetic qualities appear to have raised it to the status of artifactual icon. (photo credit 4.7)

  The Gem paper clip appears to have had its real origin in Great Britain, and the name is said by one international firm to have been “derived from the original parent company, Gem Limited.” This is supported by the Army and Navy Co-operative Society’s 1907 catalogue of the “very best English goods,” which pictures only one style of modern paper clip—a perfectly proportioned Gem, which is described as the “slide on” paper clip that “will hold securely your letters, documents or memoranda without perforation or mutilation until you wish to release them.” As early as 1908, the clip was being advertised in America as the “most popular clip” and “the only satisfactory device for temporary attachment of papers.” The ad copy went on to warn paper clip users against the use of other existing devices, whose shortcomings the Gem naturally did not share: “Don’t mutilate your papers with pins or fasteners.”