Minding Your (Claim) Language: What ‘Means For’ Means

A common idiom employed in patent claims is to define an element of the invention in terms of its function.  For example, a claim might recite an apparatus including a part ‘A’, and part ‘B’, and ‘means for fastening the part A to the part B’. 

The purpose – under Australian law – of this so-called ‘means-plus-function’ language is to define the invention as broadly as possible.  The patent specification might describe a number of different mechanisms for joining the two parts, e.g. nails, screws, and rivets. 

But there may be other ways to achieve the necessary connection, some of which the inventor has failed to think of, some of which might be unknown to the inventor, and some of which might not even have been developed at the time of filing the application.  By using broad, functional, language to define the fastening mechanism, the inventor gains the benefit of capturing all of these alternatives within the scope of the claim.

However, as a Full Bench of the Federal Court of Australian recently reminded everyone, there is a price to be paid, and a risk that is taken, through such broad claiming.  In Garford Pty Ltd v DYWIDAG Systems International Pty Ltd [2015] FCAFC 6, the court confirmed that ‘[a] claim to an apparatus for a particular purpose is a claim to an apparatus that is suitable for that purpose …. To establish infringement of such a claim it is not necessary for the patentee to establish that the apparatus is intended to be used for that purpose. The patentee need only show that the apparatus is suitable for that purpose’ (emphasis added).

Of course, what is true when assessing infringement after the date of the patent is also true when assessing novelty and inventive step based on information published before the priority date of the patent.  And, unfortunately for the patentee, Garford, the Full Court disagreed with the primary judge, and found that the prior art disclosed an element of Garford’s patent claims expressed in means-plus-function form, namely ‘a feed means [suitable] for supplying a multi-strand cable from a rotatable supply reel’ (at [124]).

There is nothing new in this, from a legal perspective.  It has long been established in Australian law that, when used in a patent claim, the word ‘for’ means only ‘suitable for’ and not, for example, specifically ‘adapted’ or ‘configured’ for the function.  It is, however, a handy reminder that claiming broadly is a double-edged sword.

But this decision also got me thinking about the implications for applicants operating in multiple jurisdictions, because the way in which means-plus-function language is treated is not the same everywhere.  The scope of claims using this format may be quite different, for example, in the US when compared with Australia, which has implications for both infringement and validity.  And while means-plus-function claiming is unobjectionable in itself in Australia, a European examiner might refuse to allow a claim in this form if it is able to be expressed in more concrete terms.

US Treatment of Means-Plus-Function

The US patent statute expressly dictates how means-plus-function claim language must be interpreted.  Specifically, 35 USC 112(f) (formerly 35 USC 112, 6^th paragraph) states that:

An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.

This paragraph was enacted in 1952 in order to overrule the Supreme Court’s decision in
Halliburton Oil Well Cementing Co. v Walker (1946) 329 US 1, which held that a claim was fatally indefinite, and thus invalid, if it defined the ‘most crucial element in the "new" combination in terms of what it will do, rather than in terms of its own physical characteristics or its arrangement in the new combination apparatus.’

The US Congress obviously considered total invalidity to be an overly harsh penalty for the use of functional language, but at the same time did not wish to allow inventors effectively to claim all ways of achieving a specified result when they had perhaps disclosed just a few – or possibly only one.

(It should be noted that while the allowance for ‘equivalents thereof’ in 35 USC 112(f) may seem quite generous, in practice it is understood that this provision is of relatively limited scope, and in particular is not the same as the test under the equitable ‘doctrine of equivalents’ which may be applied when assessing infringement.)

US v Australian Approach – Implications

The upshot of all of this is that a claim employing means-plus-function language which may be valid – although of more limited scope – in the US, would typically be of broader scope in Australia.  While this might seem to be of benefit to the inventor, as the Garford case shows it might alternatively mean that the claim is invalid in view of prior art that incidentally discloses something which, while not primarily intended to perform the claimed function nor equivalent to the mechanisms disclosed in the specification, is nonetheless suitable for the purpose at hand.

Applicants for Australian patents, whose claims have previously been examined and issued in the US, should be conscious that Australian examiners will often not conduct particularly extensive additional searching where the work has already been done by an examiner in a major foreign patent office.  It is, however, unwise to assume that a claim employing means-plus-function language is valid in Australia on the basis that it has been found valid in the US.

EPO Treatment of Means-Plus-Function

In the European Patent Office, the approach to means-plus-function language is generally similar to that in Australia, with the possible exception of elements in claims directed to computer-implemented inventions (and, formerly, the special case of so-called ‘Swiss-style’ claims for second medical uses of known substances).  Section 4.13 of the EPO Guidelines for Examination states:

If a claim commences with such words as: "Apparatus for carrying out the process etc..." this must be construed as meaning merely apparatus suitable for carrying out the process. Apparatus which otherwise possesses all of the features specified in the claims but which would be unsuitable for the stated purpose or would require modification to enable it to be so used, should normally not be considered as anticipating the claim.

… in the data-processing/computer program field, apparatus features of the means-plus-function type ("means for ...") are interpreted as means adapted to carry out the relevant steps/functions, rather than merely means suitable for carrying them out. In this way novelty is conferred over an unprogrammed or differently programmed data-processing apparatus.

My own experience is, however, that this is not the end of the story, at least with some EPO examiners.  There is a fine line between functional claiming, and ‘claiming by result’.  The latter is dealt with in Section 4.10 of the Guidelines, which explains:

The area defined by the claims must be as precise as the invention allows. As a general rule, claims which attempt to define the invention by a result to be achieved should not be allowed, in particular if they only amount to claiming the underlying technical problem. However, they may be allowed if the invention either can only be defined in such terms or cannot otherwise be defined more precisely without unduly restricting the scope of the claims and if the result is one which can be directly and positively verified by tests or procedures adequately specified in the description or known to the person skilled in the art and which do not require undue experimentation

Certainly in the case of claims using functional language in relation to computer-implemented inventions, I have encountered objections to means-plus-function type language that are, in reality, assertions that the claim is trying to define the invention by reference to a result.  Such objections are problematic, because the patent specification in such cases would typically describe the operation of the invention in some specific computing environment.  Amendments to define the invention more precisely, in order to be supported by the description, may end up limiting the scope of the claims to that particular environment.  It may or may not be possible to persuade an examiner that this is an undue restriction!

EPO v Australian Approach – Implications

Prior to the commencement of the Raising the Bar reforms in Australia, there was a significant difference between the European and Australian approach to claims-by-result.  The old ‘fair basis’ test required only general consistency between the invention disclosed in the specification and the language of the claims, and it was sufficient to describe just one way of putting the invention into effect for claims of broad scope to be valid.

The risk under this old test was that broad, functional claims that would have been acceptable in Australia might encounter objections in Europe.  Furthermore, the amendments necessary to overcome such objections could result in a significant narrowing of claim scope.

Under the current Australian law, section 40(3) requires that claims be ‘supported by matter disclosed in the specification’.  As Section 2.11.7.6A of the Australian Patent Office Manual of Practice and Procedure explains, this is intended to encompass two concepts:

1. whether the specification provides enablement across the scope of the claim; and
2. whether the claim extends beyond the contribution to the art made by the invention.

While I have yet to encounter this in practice, it seems possible that it will result in a similar approach to that taken in the EPO.

Drafting For Multiple Jurisdictions

With these kinds of differences in approach, what is an applicant to do when seeking to apply for patents in multiple jurisdictions?  This may be a particularly vexed question when proceeding via the International Application process under the Patent Cooperation Treaty, where a single patent specification must be suitable for all of the jurisdictions in which an application might ultimately proceed.

One answer would be to avoid the use of means-plus-function language altogether.  If this is possible, without sacrificing claim scope to which the applicant is legitimately entitled, then there is much to be said for it.  The goal of avoiding functional claiming must be considered from the outset, and the entire specification drafted with this in mind, with an eye to devising appropriately broad generic language to define the key features of the invention, and supporting this with as many alternative implementations as possible, falling within the generic terms.

Better yet, however, is to consider including both functional and structural claims.  While the functional claims may have different scope in different jurisdictions, there may be at least some places where they are allowable, and broader than the structural claims.  Even if they are not interpreted broadly (such as in the US), they may nonetheless provide a different scope of protection to the structural claims, and thus broaden the scope of the patent as a whole.  Structural claims, on the other hand, while potentially narrower in scope, are also less vulnerable to attack on the basis of unexpected prior art!

Conclusion – Jurisdictional Differences are a Minefield!

Unfortunately, this matter of functional claim limitation is not the only example of differences in the interpretation of language that is widely used in patent claims by practitioners across the world.

For example, for many years now Australian patent attorneys have been warning our colleagues and clients from foreign lands to be careful using the word ‘comprising’ in patent claims.  Throughout much of the world, this word is treated as a ‘term of art’ in claim drafting which, as the USPTO’s Manual of Patent Examining Procedure expresses it, ‘is synonymous with “including,” “containing,” or “characterized by,” is inclusive or open-ended and does not exclude additional, unrecited elements or method steps.’  Unfortunately, the Australian Federal Court has not always agreed. 

In Asahi Kasei Kogyo Kabushiki Kaisha v WR Grace and Co [1991] FCA 530 a judge concluded that the ‘primary natural meaning of the word “comprise”’ is ‘consists (only) of’, i.e. an exclusive and closed-ended interpretation.  The judge in General Clutch Corporation v Sbriggs Pty Ltd [1997] FCA 499 conducted an extensive literature survey encompassing various dictionaries and guides to English usage, before arriving at a similar conclusion.

Personally, I do not worry too much about ‘comprising’ these days, at least in my own patent drafting.  The overriding rule of claim interpretation is that the language used must be read in context, and it is a rare case in which it is not plainly apparent from the context that the term ‘comprising’ is intended to have an inclusive meaning.  When you have control over the entire specification from the outset, it is not difficult to ensure that the meaning of ‘comprising’ is clear from the context, or to use alternative terms if necessary.

However, it remains a live issue for applications entering Australia from overseas, where the original drafter of the specification may simply have assumed that merely by using the word ‘comprising’ the intention of inclusiveness would be apparent.  Amending to correct this may not be possible, since if the term ‘comprising’ was originally understood to be exhaustive, amending it or defining it as inclusive amounts to a broadening of the scope of the specification, which is impermissible.

The bottom line is that the more the person drafting the original patent specification knows about different laws and practices of claim interpretation around the world, the more likely it will be that the application will stand up to scrutiny in multiple jurisdictions.

Tags: Australia, Claim construction, Europe, US