In patent prosecution obviousness rejections are often resolved at the level of prior art disclosures and articulated rationales. The examiner identifies familiar components, supplies a reason to combine, and concludes that the claimed result would have followed as a predictable consequence. The response then becomes a debate over whether the references disclose each limitation or whether the rationale is sufficiently explicit. That exchange tends to accept the examiner’s underlying premise: that the system would behave in a predictable way if the references were combined. In many cases, that premise is the weakest part of the rejection. The specification is the most direct tool for testing it.
The current framework, shaped by KSR v. Teleflex and reflected in the USPTO’s examination guidance, permits a range of rationales, including predictable results, simple substitution, and “obvious to try.” MPEP § 2143 catalogs these rationales and requires that each be supported by factual findings linking the prior art to the claimed invention. At the same time, the Office must establish a reasonable expectation of success, not merely a theoretical possibility. MPEP § 2143.02 Federal Circuit decisions such as In re Magnum Oil Tools International, Ltd. and In re Van Os reinforce that conclusory reasoning is insufficient. The specification provides a structured way to challenge whether those requirements have been met.
System vs Component Level Behavior
A frequent issue arises when an examiner asserts that a claimed result would “necessarily” or “inherently” follow from a proposed combination. In practice, that assertion often rests on an oversimplified view of system behavior. Consider a claim directed to a downhole tool that induces particle separation within a helical or sinusoidal flow path. A rejection may combine a known pump with a known channel and conclude that the claimed separation effect would inherently occur. If the specification explains that separation depends on maintaining a Reynolds number within a defined band and further explains that Reynolds number is a function of curvature radius, hydraulic diameter, and fluid viscosity, the assertion of inevitability becomes difficult to sustain. The prior art may disclose each component, but it does not establish that the combined system would operate within the required regime. The argument is not that the references fail to mention the claimed feature. It is that the governing relationships described in the specification show that the feature would not arise absent additional constraints that are not taught or suggested.
Predictable Results
This same approach applies to predictable-results reasoning. The MPEP recognizes that a combination is obvious where it yields “no more than one would expect” from known elements performing their established functions. MPEP § 2141 That formulation assumes that system behavior can be approximated by summing the contributions of individual components. However, many technologies do not operate in that manner. For example, a sealing system subjected to high pressure may exhibit a non-linear response in which sealing effectiveness improves up to a threshold and then degrades rapidly due to extrusion or material fatigue. If a claim is directed to operation within that intermediate regime, the outcome is not predictable in the sense contemplated by §103 analysis. Instead, the question of obviousness depends on interactions that are not apparent from the components alone. The Federal Circuit’s reasoning in Honeywell International Inc. v. Mexichem Amanco Holding S.A. and Leo Pharmaceutical Products, Ltd. v. Rea reflects this distinction. A combination that appears straightforward at a component level may not be straightforward once system-level behavior has been considered.
Optimization
Optimization rejections are particularly amenable to specification-based rebuttal. The governing principle, reflected in MPEP § 2144.05, is that optimizing a parameter is ordinarily obvious if the prior art recognizes that parameter as affecting the relevant result. The same section makes clear that this rationale fails where the variable was not recognized as result-effective. The specification can be used to show both that the relevant variable was not identified in the art and that the claimed range represents a distinct operating regime rather than incremental improvement.
A concrete example arises in microfluidic systems for evaluating clot formation. An examiner may assert that increasing channel tortuosity is a routine design choice. If the specification explains that below a tortuosity index of approximately 2.0 the shear field is insufficient to initiate clotting, while above approximately 3.5 the flow becomes stagnant and prevents detection, the claimed range is not a matter of routine optimization. It identifies a narrow regime in which two competing effects are balanced. Unless the prior art recognizes that interplay, it does not identify the operative range. Federal Circuit precedent such as E.I. DuPont de Nemours & Co. v. Synvina C.V. confirms that a parameter cannot be treated as result-effective unless the prior art establishes a relationship between that parameter and the claimed result.
Crafting the Specification to Address Potential Obviousness Challenges
The specification can also be used to address expectation of success. The MPEP makes clear that obviousness requires at least some degree of predictability and a reasonable expectation that the combination will work for its intended purpose. Where the specification documents instability, scaling problems, or competing constraints, those disclosures can be used to show that success would not have been reasonably predictable. For example, in a high-frequency data acquisition system, the specification may describe that increasing sampling rates beyond a threshold introduces aliasing and noise amplification that degrade signal integrity. If a claim recites operation within a constrained sampling window that avoids those effects, the specification provides a basis to argue that simply increasing performance parameters would not have predictably led to the claimed system.
Unexpected results provide another avenue for using the specification as substantive support. The MPEP recognizes that results greater than expected, or qualitatively different from what the prior art would predict, are evidence of nonobviousness. MPEP § 716.02(a) Where the specification describes synergistic effects, threshold behavior, or performance reversals, those disclosures can be used to support arguments that the claimed invention is not simply an aggregation of known features.
An important practical point is that this approach does not require a declaration in every case. Where the specification already describes the relevant technical relationships, it can support factual assertions if the argument is tied to specific disclosures. Citations to particular passages, figures, or described dependencies are far more effective than generalized statements about how the invention operates. PTAB panels routinely distinguish between unsupported attorney argument and arguments grounded in the intrinsic record with only the latter treated as evidence.
The implications for drafting are direct: A specification that merely lists components and exemplary embodiments provides limited leverage in responding to §103 rejections. Conversely, a specification that explains how system behavior changes with operating conditions, identifies limits and failure modes, and describes interactions between variables creates a record that can be used to test the assumptions embedded in a rejection. Moreover, this does not require extensive experimental data. Even qualitative descriptions of how performance shifts across a range of conditions can be sufficient to support later arguments. Used in this way, the specification allows the applicant to move the analysis away from generalized statements about what is “predictable” and toward whether the examiner’s reasoning is consistent with the actual behavior of the claimed system. That shift is often where §103 disputes are resolved.
For companies seeking stronger patent protection, a well-developed specification can do more than support filing, it can create meaningful leverage during prosecution when responding to obviousness rejections. Contact Conley Rose to discuss how your organization can strengthen its patent drafting strategy to better support long-term prosecution objectives.