January 15, 2018
Transparent Contract Componentssharper tools, finer toolmarks
Continuing from Contract Components and Lexical Scope for Contracts
I got started on the contract-components problem by giving myself permission to kick orthodoxy to the curb. That freedom showed me that reusable components of contract language are possible. But all of my approaches to date suffered a common flaw: Reviewing a composed document, it would be very clear something newfangled and strange was going on.
How do we make contracts out of components, without calling attention to how our contracts get made? Two tools.
1: Term Substitution
Consider the example component from my first post, a software license grant that uses, but does not define, Licensor, Licensee, and Program:
The Licensor grants the Licensee a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable copyright license to reproduce, prepare Derivative Works of, publicly display, publicly perform, sublicense, and distribute the Program and such Derivative Works.
“Derivative Works” means any work that is based on (or derived from) the Program and for which the editorial revisions, annotations, elaborations, or other modifications represent, as a whole, an original work of authorship. Derivative Works do not include works that remain separable from, or merely link (or bind by name) to the interfaces of, the Program and Derivative Works of the Program.
Including that component in an agreement that defines Vendor, Customer, and Software, we can substitute agreement terms for component terms:
The Vendor grants the Customer a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable copyright license to reproduce, prepare Derivative Works of, publicly display, publicly perform, sublicense, and distribute the Software and such Derivative Works.
“Derivative Works” means ….
This solves the problem of passing terms down to components.
Compared to prior proposals, it’s a bit more strict. Every term in a component must be identical to another term in the incorporating document, which means it has to be defined to begin with. It also sacrifices perfect reproduction of components, without changes. This approach modifies components, but in a predictable way.
2: Conflict Resolution
We are still working under traditional definition-scope rules, which apply definitions throughout the documents where they appear. That means using multiple components can produce an overall agreement with multiple definitions of the same term. Definitions in the containing document can conflict with definitions in components. Definitions in components can conflict with definitions in other components.
We need a rule to resolve those conflicts:
Where this document uses a term defined more than once, the definition nearest the use takes precedence. To find the nearest definition, look for a definition in the organizational subdivision where the term is used, then in the subdivision containing that subdivision, then in the subdivision containing that subdivision, and so on.
(To see my latest revision of this language, check it out on commonform.org.)
This rule follows my proposal in Lexical Scope for Contracts by reusing the organizational structure of the document itself to create boundaries for scopes. It differs from prior rules by doing no more than necessary. The rule only applies to, and addresses, situations where a term is defined more than once. It does not change the general rule of definition scope in any way.
This approach isn’t fundamentally different from what I described before. It merely picks trade-offs netting out to a more familiar reading experience. Products of this system can probably “pass” in the practice of law today. Products of prior systems would take some explaining.
Newfound readability will break down in some parasitic cases. If there are lots of definition conflicts, readers will get tired of applying the nearness rule to resolve them. But insofar as conflicts are rare to begin with, the cost is zero. We can even use software to tell whether our compositions have conflicts, and omit the resolution rule if the document is clean without it.
Your thoughts and feedback are always welcome by e-mail.
more articles — revision history — back to top