Patent Task§ JOURNAL · Tutorial
Antecedent Basis in Patent Claims: Examples and Practical Corrections
An antecedent-basis review asks whether a reader can identify the element or feature that a claim refers to. “A sensor” followed by “the sensor” is straightforward when there is one relevant sensor. The same reference becomes less helpful when the claim introduces two sensors and never says which one controls the next operation.
USPTO MPEP § 2173.05(e) discusses missing and ambiguous antecedents. It also expressly recognizes that a missing explicit antecedent does not always make a claim indefinite: the question includes whether a skilled reader can reasonably ascertain the scope. A terminology warning is therefore a reason to inspect the claim, not an automatic finding that a patent is invalid.
This page develops one part of claim review. Use the independent and dependent claims example for claim structure and the US drafting guide for the wider description-and-support review.
Establish the technical identities first
All examples below are fictional teaching material created for this guide. They are not granted claims, client material or samples from a Patent Task drafting task.
The disclosed assembly contains:
| Source fact | Component or relationship |
|---|---|
| S1 | An inlet temperature sensor measures inlet-air temperature. |
| S2 | A separate outlet temperature sensor measures outlet-air temperature. |
| S3 | A fan is controlled by a controller. |
| S4 | The controller activates the fan when the measured outlet-air temperature exceeds a threshold. |
| S5 | An optional warning indicator is connected to the controller, which illuminates it while the fan is activated. |
No threshold value, cooling-performance result, indicator technology or sensor construction is supplied. The examples concern identity and reference, not novelty or an appropriate protection strategy.
Example 1: a reference with no introduced component
Consider this isolated draft fragment:
An air-monitoring assembly comprising a fan and a controller configured to activate the fan when a temperature measured by the outlet temperature sensor exceeds a threshold.
The fragment refers to “the outlet temperature sensor” without introducing that sensor. The disclosure contains the component, so one possible correction for this exercise is to introduce it explicitly:
An air-monitoring assembly comprising an outlet temperature sensor configured to measure an outlet-air temperature; a fan; and a controller configured to activate the fan when the outlet-air temperature measured by the outlet temperature sensor exceeds a threshold.
The change makes the sensor's identity explicit. It also adds that sensor to the listed combination. A practitioner should assess the resulting scope; repairing a reference is not a reason to add whichever component an automated checker suggests.
This is an illustrative claim fragment. It does not address every feature or drafting decision for the hypothetical assembly.
Example 2: two introduced sensors, one ambiguous reference
A different draft could introduce both sensors and then say:
... an inlet temperature sensor; an outlet temperature sensor; and a controller configured to activate a fan when a temperature measured by the sensor exceeds a threshold.
The problem is not that a sensor has never been mentioned. There are two candidates. Replacing “the sensor” with “a sensor” would not identify which measurement governs the operation.
For S1–S4, a clearer teaching claim is:
1. An air-monitoring assembly comprising an inlet temperature sensor configured to measure an inlet-air temperature; an outlet temperature sensor configured to measure an outlet-air temperature; a fan; and a controller configured to activate the fan when the outlet-air temperature measured by the outlet temperature sensor exceeds a threshold.
This wording follows S4 by identifying the outlet measurement. Changing it to the inlet measurement would alter the disclosed control condition, not merely improve grammar.
The example includes the inlet sensor to demonstrate two distinct identities. Its inclusion is not a recommendation that every real independent claim should contain every disclosed component.
Example 3: introduce a new element in a dependent claim
A dependent claim may introduce a new component while further limiting the parent combination. The USPTO utility application guide explains that dependent claims refer back and add limitations. The new component does not have to appear in the parent merely so that it can be introduced properly in the dependent claim.
Using claim 1 above and optional fact S5:
2. The air-monitoring assembly of claim 1, further comprising a warning indicator connected to the controller, wherein the controller is configured to illuminate the warning indicator while the fan is activated.
The phrase “a warning indicator” introduces the additional component. “The warning indicator” then refers to that component. “The controller” and “the fan” refer to elements inherited from claim 1.
Contrast this with an opening that says “wherein the warning indicator is illuminated” without introducing an indicator anywhere in the dependency chain. Do not assume an omitted component must be inserted into claim 1; decide where it belongs in the intended claim structure and confirm its disclosure support.
Keep a terminology register through revisions
| Term being reviewed | Introduction or inherited source | Later reference | Review result in this example |
|---|---|---|---|
| Inlet temperature sensor | Claim 1 | Inlet measurement | Separate identity; not the measurement named in the fan-control condition |
| Outlet temperature sensor | Claim 1 | Outlet measurement in the control condition | Matches S4 |
| Fan | Claim 1 | “the fan” in claims 1 and 2 | Same component through the dependency |
| Warning indicator | Claim 2 | “the warning indicator” in claim 2 | New element introduced before reuse |
| Display | Not introduced | If added by a later edit | Determine whether it means the warning indicator or a different component |
Do not automatically replace “display” with “warning indicator.” It could refer to a genuinely different feature in a later source version. Resolve the technical identity first, then adjust the words and review the affected claims.
Use a focused review pass
- Read the claim and its dependency chain together.
- Mark introduced components and attributes.
- For each later reference, identify the intended earlier element.
- Check competing candidates, such as two sensors or two surfaces.
- Compare any proposed correction with the disclosure and intended scope.
- Recheck dependent claims and description terminology after the edit.
Matching words alone is insufficient. Conversely, the MPEP recognizes that identical wording in the claim and specification is not always required. Keep this reference check separate from the further questions of written description, enablement and prior art.
Bring the reviewed terminology into the US drafting workspace, and inspect generated or revised claims against your source facts. The claim support worksheet can record a limitation, its source and the correction still needed.
When an Office Action raises a § 112(b) antecedent-basis rejection
Start with the examiner's exact concern and the claim version addressed by the action. Confirm the current pending claims and their dependencies, then inspect the possible referents alongside the application-as-filed specification and drawings. A clarity question under § 112(b) and a disclosure-support question under § 112(a) require distinct review; changing an article or component name does not resolve both automatically.
OA Review's patent Office Action response-preparation checklist provides a general method for assembling that record, preserving open questions and handing them to the responsible practitioner. OA Review is a separate product built by the same team as Patent Task. Opening the resource does not carry your documents or account into that product.
Keep any proposed correction separate from pending claim text and locate its support in the actual as-filed record. Newly generated drafting material cannot retrospectively supply that support. The practitioner decides whether and how to amend or respond.
Does every use of “the” require an earlier identical phrase?
Not as a mechanical rule. Read the claim in context and consider inherent features and reasonably identifiable references. Explicit, consistent introductions can make review easier, but a word-matching score is not a legal conclusion.
Can a new component appear for the first time in a dependent claim?
Yes. The indicator example introduces a component and adds a limitation while retaining the parent combination. Check the full dependency and disclosure rather than requiring every noun to originate in claim 1.
Does correcting antecedent basis make a claim patentable?
No. A clearer reference does not establish sufficient disclosure, novelty, nonobviousness or a useful scope of protection.