Provisional Application for a Patent

Jed Margolin

I am not a Patent Attorney or Agent. These are my own opinions.

Sometimes a Provisional Application for a Patent is referred to as a Provisional Patent Application (even by the USPTO) which leads some people to conclude that it is an application for a Provisional Patent. It isn't. There is no such thing as a Provisional Patent. When you see or read a news report or press release that refers to a company having applied for or obtained a Provisional Patent it means that someone doesn't know what they are talking about. It could be the reporter or the company representative. When it is a company's president making the claim, either they have  been misquoted or they are just plain ignorant.

A Provisional Application for a Patent is a disclosure that can be used establish the priority of your invention. It's not the only way, but that requires some explanation.

The United States gives priority to the inventor who is the First-to-Invent, as opposed to the rest of the world, which uses a First-to-File system. There is the often-told story about how Alexander Graham Bell got to the Patent Office a few hours before Elisha Gray on the morning of February 14, 1976, and for that reason got the patent for the telephone, suggesting that the First-to-File system was in effect at that time. All the evidence says otherwise. More here.

But how do you prove that you were the first to invent something? The Patent Office will not simply take your word for it.  The standard method for establishing the conception of an invention is through the use of witnessed, permanently bound, and page-numbered laboratory notebook or by notarized records.

The problem is that very few scientists and engineers are familiar enough with patent law to put into their lab notebooks the information that is needed to legally qualify as a complete disclosure.

As a result, most patent interference actions are won by whoever was the first to file.

Another complication is that in the U.S. you have up to a year to apply for a patent for an invention after you publish the details of your invention, sell it, or even just demonstrate it to the public. In most of the rest of the world you cannot get a patent on an invention after you have made the details public, etc.

Through the Patent Cooperation Treaty (PCT) with various nations and the more recent GATT Uruguay Round Agreements, patent applications filed in one of the signatory countries are given a priority date based on the date they filed a patent application in their respective countries. However, that does not apply if you have published the details of the invention, etc.  before applying for a U.S. patent.

This is where the Provisional Application for a Patent {sound of trumpets} comes to the rescue.

The signatory nations of the PCT and GATT Uruguay Round Agreements recognize a Provisional Application for a Patent as a real patent application.

You can file a Provisional Application for a Patent, then publish the details of the invention, demonstrate it, sell it, and you have up to a year to file a Non-Provisional (Real) Patent Application, while preserving your rights to file patent applications in foreign countries, which is what you have to do because there is no such thing as an International Patent.

A Provisional Application for a Patent is simpler to prepare and costs less to file than a Non-Provisional Patent Application.  It also can be a trap for the unwary. More about that in a moment.

First, a short rant.

I wish they wouldn't call a Real Patent Application a Non-Provisional Patent Application. A  product should be called by what it is, not by what it is not. Calling it by something that it is not carries with it a negative connotation and is very poor product positioning. (The TV industry eventually renamed Non-Interlaced Scanning as Progressive Scanning.) Instead of Non-Provisional Patent Application I suggest Regular Patent Application. (Real Patent Application would be even better, but that would suggest that a Provisional Application for a Patent is not a real patent application, and we have all agreed to pretend that it is.)

After filing a Provisional Application for a Patent you have one year to file a Non-Provisional Application and be able to claim priority for the filing date of the Provisional Application. If you wait more than one year you lose that right. You can still file a Non-Provisional Application but you cannot claim the priority date of the Provisional Application.

In order to discuss the traps for the unwary that I mentioned earlier, I need to explain something about patents. If you already know this stuff, then skip to the next section.

Some people read the Abstract of the patent, think it is the invention, and are outraged, because they feel it is so broad.

The Abstract is not the invention. In fact, the Abstract has no legal meaning for the patent. The only purpose of he Abstract is to provide search terms.

Some people read the Specification of the patent, think everything in it is the invention, and are outraged because there is so some much in it that everybody already knows.

The legal definition of the invention is in the Claims. The claims are written in a form of legalese with very particular requirements. Think of it as a language with a rigid grammar. An analogy can be made with Latin. Although Latin has a rigid grammar, within the framework of the grammar there is considerable flexibility in expressing ideas. Ideas can be expressed in ways that are murky and difficult to understand. Some expressions are simple and clear. Some are even elegant. An analogy that more people will understand is a programming language like "C."

The purpose of the Specification is to provide enough detail to enable a Person having Ordinary Skill in the Art (POSITA) to build the invention without undue experimentation.

The Specification also serves as a "dictionary" for the claims.

Originally, the claims stood almost alone, and using the Doctrine of Equivalence, could be interpreted very broadly in ways not even suggested by the Specification.

However, due to a relatively recent Supreme Court decision (Festo), the Specification also serves to limit the claims. How much the Specification is used to limit the claims has not been totally worked out. There is a good explanation of the Festo decision in the article Patent Pending...Pending...Pending The Evolution of Equivalents: Festo Corp. vs. Shoketszu Kabushiki Co. by Karly Stoehr.  Here are some comments on the Festo decision.   Here are some more comments.  Festo Corporation, aka Festo USA, is a subsidiary of The Festo Group, a large global company headquartered in Germany.  There are 52 independent Festo companies with over 250 branch offices. The other party to the lawsuit, Shoketszu Kabushiki, does not seem to have a Web presence.

Many people consider Claims to be a necessary evil because they are so difficult. The reason they are so difficult is that they define the invention. It is likely you will not fully understand exactly what your invention is until you are satisfied with the claims. At that point you need to go back and make sure the Specification properly supports the claims.

If you are interested in claims read some patents. Read lots of patents.

(Some people read only the preamble to a claim, think it is the invention, and are outraged. You need to accept that some people insist on being outraged by patents that belong to somebody else.)


I won't repeat everything the Patent Office has to say about the Provisional Application for a Patent.

Here are some highlights and pitfalls.

  • A Provisional Application does not require a claim.

  • When the Provisional Application was introduced, the rules in the section covering Real Patent Applications continued to state that at least one claim was required, leading to the fear that a Provisional Application without a claim might not count as a real patent application by other countries. As a result, many patent practitioners included one broad claim in Provisional Applications. Since then, the Rules have been amended to state that a claim is not required in a Provisional Application in order for it to be recognized as a patent application under PCT and GATT. However, since it has not been tested in a court, many practitioners continue to include one claim in a Provisional Application. (The Rules never said it has to be a good claim.

    Here are the Pros and Cons on including a claim in a Provisional Application:

    Pro: If you include a claim you don't have to worry that somewhere down the road a court somewhere in the world will rule that a patent application has to have at least one claim.

    Con: If your claim is narrower than you are entitled to, you will not be permitted to broaden it during the prosecution of the patent.

    But here is the trap.

    As I have previously stated, "It is likely that you will not fully understand exactly what your invention is until you are satisfied with the claims. At that point you need to go back and make sure the Specification properly supports the claims."

    If you have filed a Provisional Application without having written the claims, or at least having a really good idea of what your claims are going to be, the Specification may not support the claims you would really like to have. Since you are not permitted to add new material to the Specification, you are hosed.


        Ed Suominen has written an excellent article "Re-Discovering Article 1, Section 8 - The Formula for First-to-Invent." where he presents the case that the clause of the U.S. Constitution that is the basis for the U.S. Patent system requires that it be a First-to-Invent System.  Article 1 Section 8, gives Congress the authority:

    Clause 8: To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries;
    Mr. Suominen shows how the various Patent Acts and legal cases since then have interpreted Inventor to mean First and True Inventor . In the process it is clear that we have always had a First-to-Invent system.

    The article was published in the Journal of the Patent and Trademark Office Society (JPTOS) in the September 2001, Volume 83, No. 9 issue. Mr. Suominen has posted the article on his web site ( at

    I have also found additional supporting evidence.

    From the article Inventing the Telephone from the AT&T web site at

    There's a well known tale that Bell beat another inventor, Elisha Gray, to the patent office by a few hours. While true, it's not the whole story. Bell filed a patent application, a claim that I have invented. Gray, on the other hand, filed a caveat, a document used at the time to claim I am working on inventing. Priority in American patent law follows date of invention, not date of filing. Still, filing first helped Bell avoid a possible costly and time-consuming dispute. The U.S. Patent Office issued patent #174,465 to Bell on March 7, 1876.
    However, AT&T may have a high threshold of what constitutes a "costly and time-consuming dispute."

    The Editorial Review of the book The Telephone Patent Conspiracy of 1876:  The Elisha Gray-Alexander Bell Controversy and Its Many Players  by A. Edward Evenson notes:

    From Book News, Inc.
    Not long after the US Patent Office confirmed Alexander Graham Bell as the inventor of the  telephone, a flood of lawsuits washed away all certainty of his claim. The government would eventually try to annul Bell's patent. This work chronicles the rivalry between Bell and inventor Elisha Gray, drawing on court testimony, contemporary accounts, government documents, and the  participants' correspondence. The author is an industrial engineer. Book News, Inc.®, Portland, OR.
    (You can go to and search for "Telephone Patent Conspiracy" to see more reviews.}

    Hopefully, AT&T's assertion about First-to-Invent was more accurate.

    This evidence may be more persuasive.

    From The Story of the U.S. Patent and Trademark Office, available at

    "This is an incomplete draft of a digital version of The Story of the U.S. Patent and Trademark Office, with significant additions. Special thanks is given to The Office of Public Affairs, USPTO. This chronological listing of the history of the Patent and Trademark Office has been out of print since the printing done for the Bicentennial in 1990, and that was the 1988 version. "
    {That was to establish the credibility of the source.}

    From the entry for 1860:

    Philip Francis Thomas, eighth Commissioner of Patents, February 16, 1860 to December 10, 1860. He later becomes Secretary of the Treasury.
    Commissioner Thomas sets aside one of his ablest examiners to hear and determine interferences. An "interference" is a proceeding for determining who is the original inventor when two or more seek a patent for the same invention at the same time. The one who made the invention first is entitled to the patent.
    If it had been a First-to-File system the question would have been who was the first to file, not who made the invention first.

    You only need to determine who the original inventor is if it is a First-to-Invent system.

    If it was a First-to-Invent system in 1860 it was almost certainly a First-to-Invent system in 1876 when Bell beat Gray to the Patent Office.

    However, it's also possible that neither of them was the original inventor of the telephone. Here is the case for Antonio Meucci.

    Here is the USPTO's map of PCT Contracting States.

     PCT Countries           Non-PCT Countries 


    Here is information on the PCT from the USPTO.

    Here are some Basic Facts about the Patent Cooperation Treaty (PCT).

    Here is more information about the PCT from the World Intellectual Property Organization (WIPO).

    And here is the list of PCT contracting states from WIPO.

    Jed Margolin
    San Jose, CA
    June 10, 2004

    Please send comments here