Trade Secrets

Kadant v. Seeley: Reverse Engineering as a Complete Answer

A Northern District of New York court denied a trade-secret injunction where a former employee's new employer plausibly reverse-engineered publicly available parts—and the plaintiff could not prove its specifications were secret or improperly taken.

November 11, 2025
An engineer measuring a metal industrial nozzle with calipers at a workbench
Kadant v. Seeley shows reverse engineering of lawfully available products defeating a misappropriation claim at the injunction stage. Shutterstock
Educational content, not legal advice. This article explains general legal concepts. It does not create an attorney–client relationship. For your specific situation, consult a licensed attorney.

In Kadant, Inc. v. Seeley Machine, Inc., No. 02-CV-1568 (DNH/RFT), 244 F. Supp. 2d 19 (N.D.N.Y. Jan. 30, 2003), District Judge David N. Hurd of the United States District Court for the Northern District of New York drew one of the cleaner lines in trade-secret law between misappropriation and lawful competition. A papermaking-equipment maker watched a departed employee help a rival field a competing product line with startling speed, and inferred theft. The defendants answered that they had simply reverse-engineered parts that were freely available in the market and unprotected by any patent. At the preliminary-injunction stage, the court sided with the defendants on the trade-secret claim: a plaintiff who cannot show that its information was secret, or that the defendant’s products were obtained by improper means rather than reverse engineering, is not entitled to an injunction. The case is a working illustration of reverse engineering operating as a defense.

At a glance

  • Case: Kadant, Inc. v. Seeley Machine, Inc., No. 02-CV-1568 (DNH/RFT)
  • Court: U.S. District Court for the Northern District of New York
  • Judge: District Judge David N. Hurd
  • Decided: January 30, 2003
  • Posture: Motion for a preliminary injunction
  • Governing law: New York trade-secret law (common law, informed by the Restatement (First) of Torts § 757)
  • Holding: Preliminary injunction granted on the trademark claim but denied on the trade-secret, breach-of-contract, and fiduciary-duty claims; the plaintiff failed to show a likelihood of success because it did not establish that its design specifications and customer data were protectable secrets, or that the defendants’ competing products were improperly obtained rather than reverse-engineered from lawfully available products
  • Significance: A concrete application of the principle that reverse engineering of lawfully acquired products is a proper means, and that the misappropriation plaintiff bears the burden of negating it

Kadant, through its AES division in Queensbury, New York, had for decades manufactured products that clean and condition papermaking machines and filter the water used in the process. Stephen Corlew, a former Kadant employee, went to work for Seeley Machine in late April 2002, where he helped develop and market a new line of competing products for the pulp and paper industry. Kadant alleged that the new line appeared too quickly to have been developed honestly, and that Corlew must have taken Kadant’s confidential design specifications and customer databases. It moved for a preliminary injunction to shut the new line down.

The plaintiff’s speed-based inference

Kadant’s central argument was circumstantial and intuitive. Developing a competing line of nozzles and related equipment, it contended, requires substantial time, money, and technical skill; reverse engineering in particular is slow and laborious. Kadant put a number on it, asserting that it would take the defendants roughly 1.7 years to reverse-engineer its full nozzle line. Because the defendants brought a competing line to market far faster than that, Kadant reasoned, they must have skipped the honest labor and instead used Corlew’s misappropriated specifications.

The logic is familiar in employee-mobility cases: rapid replication is offered as proof of theft. But it is an inference, not direct evidence, and it carries weight only if two predicates hold—that the underlying information was actually a protectable trade secret, and that there was no lawful path to the same result quickly enough to explain the defendants’ speed. The court found both predicates unproven.

Reverse engineering as a proper means

The defendants’ answer went to the heart of trade-secret doctrine. They asserted that the products in the new line were reverse-engineered from existing products that were freely available in the public domain and unprotected by published patent applications, in-force patents, or trade secrets. That assertion, if credited, is dispositive, because reverse engineering of a lawfully obtained product is a quintessential “proper means” of acquiring the know-how embodied in it. A secret that can be ascertained by examining a product sold on the open market is, to that extent, not much of a secret at all; the law has consistently refused to let a manufacturer use trade-secret doctrine to bar competitors from studying products the manufacturer itself put into commerce.

The court held that Kadant had not carried its burden to show a likelihood of success. It had not established that its design specifications and customer databases qualified for trade-secret protection in the first place, and it had not shown that the defendants’ products were improperly obtained as opposed to reverse-engineered. On a preliminary-injunction motion, where the movant must demonstrate a likelihood of success on the merits, those failures were fatal to the trade-secret claim. The defendants were left free to resume developing, marketing, and selling their competing products.

Notably, the court did not let the defendants off entirely. It granted a preliminary injunction on Kadant’s trademark claim, enjoining the defendants from using the acronyms “AES” or “APS”—or any other three-letter acronym beginning with “A” and ending with “S”—because of the likelihood of confusion with Kadant’s marks. The split outcome is instructive: the same set of facts can support a strong claim on one theory (source confusion in branding) and a weak one on another (theft of technical secrets), and courts will sort them claim by claim.

The burden problem and the limits of inference

Kadant is, at bottom, a case about who must prove what. The plaintiff wanted to use the defendants’ speed to shift attention away from the threshold questions—secrecy and improper acquisition—but the court kept the burden where the law places it. A misappropriation plaintiff must identify protectable secrets with enough specificity to show they are not generally known or readily ascertainable, and must demonstrate that the defendant acquired or used them by improper means. Where the defendant offers a plausible lawful explanation—reverse engineering of marketed products—the plaintiff cannot prevail merely by asserting that the lawful path would have been slower than the defendant’s results suggest. It must produce evidence that the lawful path was not in fact taken, or that the information was not available through it.

The “1.7 years” figure illustrates the danger of resting on such estimates. A reverse-engineering timeline is only as good as its assumptions about who is doing the work, what tools and prior knowledge they bring, and how much of the product is self-revealing once disassembled. An estimate built for one purpose can collapse against a defendant who shows that much of the design was visible on inspection or already public.

Open questions

The decision resolves a preliminary-injunction motion, not the merits, and several issues were left for later development. First, the court did not finally decide whether any of Kadant’s specifications or its customer database qualified as trade secrets; a fuller record could conceivably support protection for some discrete information, particularly compiled customer data, even if the hardware was reverse-engineerable. Second, the opinion does not resolve the breach-of-contract and fiduciary-duty theories on the merits—conduct that is not trade-secret misappropriation can still breach a non-disclosure covenant or a duty of loyalty, and those claims survived the injunction denial. Third, Kadant leaves open how courts should treat a defendant who possessed misappropriated materials but could also have reverse-engineered the result: when both a tainted and a clean path exist, allocating proof of which one the defendant actually used is a recurring and unsettled problem.

Implications

  • Reverse engineering of lawfully acquired products is a defense. A competitor who studies and replicates a product sold on the open market has used a proper means; trade-secret law does not bar it.
  • The plaintiff carries the burden on secrecy and improper means. Speed of replication is not, by itself, proof of theft; the movant must show the information was a protectable secret and that it was taken improperly.
  • Identify and isolate genuine secrets. Information that is readily ascertainable from a marketed product is weakly protected; plaintiffs should focus on truly non-public material, such as discrete customer or cost data, rather than reverse-engineerable hardware.
  • Plead alternative theories. As the trademark injunction shows, the same facts may fail under trade-secret law yet succeed under trademark, contract, or fiduciary-duty theories; claims should be evaluated independently.
  • Timeline estimates are vulnerable. Reverse-engineering duration estimates depend heavily on contestable assumptions and rarely substitute for direct evidence of misappropriation.

Frequently asked questions

Why did Kadant lose its trade-secret claim but win on trademark? The court analyzed each claim on its own merits. Kadant could not show its technical specifications were protectable secrets or that the defendants took them improperly rather than reverse-engineering available products, so the trade-secret injunction was denied. The trademark claim turned on a different question—likelihood of consumer confusion from the defendants’ use of similar three-letter acronyms—which Kadant did establish.

Is reverse engineering always a defense to trade-secret misappropriation? It is a proper means only when the product reverse-engineered was lawfully acquired, typically by ordinary market purchase. It is not a defense where the defendant obtained the item or the underlying information by improper means or in breach of a duty, such as a contractual non-use restriction.

What does Kadant teach trade-secret plaintiffs? Do not rely on the inference that fast replication proves theft. Identify specific, genuinely secret information; show it could not have been readily reverse-engineered or otherwise lawfully obtained; and marshal direct evidence that the defendant actually used improper means.

Authorities and sources