Contract Interpretation Grievances

Contract interpretation grievances sit at the intellectual centre of labour arbitration. They rarely turn on disputed facts. They turn on the meaning of the words in the agreement, and that meaning is rarely obvious in isolation. A good interpretation argument reads the article in dispute against every other article in the agreement, the parties' history, and the arbitral case law on similar language.

Canadian arbitrators work through a well-established interpretive framework. The starting point is the plain and ordinary meaning of the disputed words, read in the context of the whole agreement — the whole-agreement canon — so that no provision is read in isolation and no clause is rendered redundant. Where the language is clear, that is usually the end of the analysis. Where the language is ambiguous, arbitrators consider extrinsic evidence: past practice, bargaining history where admissible, memoranda of agreement, letters of understanding, and the body of arbitral case law that has interpreted similar language in similar contexts. The exercise is less about finding the "right" answer than about locating the most defensible interpretation given everything the parties have actually agreed to and done.

Past practice is frequently the decisive factor when the language permits more than one reading. Where the employer and the union have applied a provision consistently over the life of the agreement — or several successive agreements — and neither side has objected, that practice can harden into a binding interpretation. Estoppel is the doctrine that carries this through: a party that has relied on the other side's consistent interpretation to its detriment can prevent the other side from changing the interpretation unilaterally. Estoppel is typically a union tool on interpretation grievances, but the doctrine is available to employers as well, and arbitrators have applied it in both directions when the factual record supports it.

Interpretation files are also where multi-agreement comparison matters most. Many employers — especially in the public sector — have several bargaining units with similar but not identical language on the same subject. A scheduling article in the nursing agreement may read differently from the same subject in the support-staff agreement, and the difference is sometimes deliberate and sometimes historical. Sertus supports side-by-side comparison across agreements so counsel can see the drift before taking an interpretation position on any one of them. The decision library rounds out the workflow: counsel can pull comparable arbitration awards on the same language and the same subject in natural language, so the interpretation argument is grounded in actual Canadian arbitral practice rather than in an abstract reading of the clause.