Defensive Publication: Why I File a Patent Application, Let It Publish, Then Abandon It

In my 20 years of practice as a patent attorney and litigator, defensive publication is something I recommend often. But my approach differs from the textbook one. I rarely tell clients to publish in a journal or a disclosure database. Instead, I tell them to file a patent application, wait for it to publish, and then abandon it. Here is why, and when.

What is a defensive publication, really?

A defensive publication puts a technology into the public domain so that it becomes prior art. Once something is prior art, nobody can later obtain a patent on the same invention, because it is no longer new; close variations also struggle to clear the inventive step bar.

It is a trade. You give up the right to exclude others, and in return nobody else gets that right either. You cannot sue anyone with a publication, but nobody can sue you with a patent on what you published.

Which companies should consider defensive publication?

The real purpose of a patent is to protect your market when you need to, which in practice means being willing and able to enforce it. If a company knows from day one that it will not enforce in a particular country, then filing there, refining the claims and paying ongoing fees is spending with no return.

The clients I most often give this advice to are German companies operating in China. Because damages awards are too low in their view, many German companies are unwilling to enforce patents in China. What they actually worry about is not "can I sue someone?" but "will a competitor patent this first and use it against me?" For that concern, a defensive publication fits better than a full patent.

So the question I ask clients is simple: if you had a patent on this, would you, and could you, actually enforce it in this market? If the honest answer is no, defensive publication deserves a serious look.

A patent earns its keep through enforcement. Where a company will not enforce, what it usually needs is not a patent but a public document that stops anyone else from getting one.

This is not legal advice; consult a qualified patent attorney for your situation.

Why publish through a patent application instead of a journal?

To get a better result, I generally do not recommend any print or electronic journal. I recommend disclosing in the form of a patent document and abandoning the application once it has been published.

The reasoning is practical. A defensive publication only works if a future examiner, reviewing someone else's application, can find it. Patent documents are where examiners search first: they share a standard format, carry classification codes, have clear filing and publication dates, and are indexed in patent offices' search systems worldwide. A journal article may be just as valid as prior art in legal terms, but whether an examiner ever sees it depends on where and how it is indexed. A disclosure no one finds offers little protection.

Patent-form disclosures also have a timing advantage that journals lack. Several major systems treat a published earlier application as prior art from its filing date, not just its publication date, at least for novelty purposes:

• China: under Article 22(2) of the Patent Law, an earlier application by anyone, published after a later application's filing date, can destroy the later application's novelty (the "conflicting application" rule).
• Europe: under Article 54(3) EPC, earlier European applications published later form part of the state of the art for novelty only.
• United States: under 35 U.S.C. 102(a)(2), a published U.S. application is prior art as of its effective filing date.

A journal article only counts from the day it appears.

How well does a defensive patent application need to be drafted?

This is where companies most often waste money.

When the goal is publication, there is no need to spend on drafting quality. Claim strategy, careful generalisation, room for amendment during prosecution: all the work that makes a granted patent valuable can be skipped, because you never intend to get a grant.

What you cannot skip is disclosing all the necessary technical content. "Necessary" means enough for a skilled person to carry it out: the structure, the steps, how key parameters are chosen, and the workable alternatives. Anything you leave out does not become prior art, and someone else may still patent it. The document can be plain; it cannot be incomplete.

One more point: what you publish must be technology you actually developed. The aim is to put real engineering into the public domain, not to pad a document so it looks substantial.

A defensive filing doesn't need polished claims, but it does need complete technical content. Published this "unconventional" way, it often costs less for the protection it gives.

What is the practical sequence, and when do you abandon?

In China, I usually take this route with an invention application:

1. Prepare the document, focusing on the description so that all necessary technical content is there. The claims only need to satisfy formal requirements.
2. File and wait for publication. Under Article 34 of the Chinese Patent Law, an invention application is published after eighteen months from filing, and the applicant may request earlier publication. If you want prior art quickly, request early publication.
3. Abandon only after publication. The order matters. Withdraw before publication and the document never becomes public. After publication, simply stop prosecuting, for example by not requesting substantive examination, and the application lapses.

Note that Chinese utility model applications are only made public when they are granted, so for this purpose an invention application is the usual vehicle. Publication timing and abandonment mechanics differ in other countries, so check locally before relying on this sequence.

What mistakes should you avoid?

• Publishing core technology you will want to enforce. Publication cannot be undone; even you cannot later patent what you have published. Technology you plan to enforce should be patented normally.
• Publishing what cannot be reverse-engineered. Some processes and formulations cannot be worked out from the product. These are usually better kept as trade secrets; publishing them hands competitors something they could never have obtained otherwise.
• Publishing an outline instead of an implementation. If the key implementation details are missing, a competitor can still patent the specifics.
• Abandoning before publication. It is a simple procedural mistake, but it wipes out the whole exercise.

The bottom line

Defensive publication is not a fallback for inventions that "can't be patented". It is a deliberate commercial choice: in markets where you will not enforce, it secures your own freedom to operate at minimal cost.

Before you publish, check what is already out there. The search tool below helps you see the existing prior art landscape. Whether and how to proceed should still be confirmed with a qualified patent attorney.