This term does not exist (to my knowledge), hence just use the term pedagogical modelling language (PML) for convenience to aid in further explanations to refer to a language that is used for modelling pedagogy in a manner that can be reasoned upon to come to decisions around the learning design. That is, the PML is used to aid in the production of [[Learning design|learning designs]] that can be represented formally by an [[Educational modelling language|educational modelling language (EML)]], such as [[Orchestration graphs|orchestration graphs]]. In my thought about [[The automated derivation of design-related contextual pedagogical theory|the automated derivation of design-related contextual pedagogical theory]], the representation of the derived theory would be in the form of a pedagogy modelling language. [[Educational modelling language|EMLs]] have pedagogy embedded within them, where the language mediates the pedagogical thought processes of the learning designer, such as the pedagogical edges describing relationships between activities in [[Orchestration graphs|orchestration graphs]]. A collection of [[Pedagogical patterns|pedagogical patterns]] is expressed through [[Educational modelling language|EMLs]], however here I propose an alternative where we try to capture and express the laden pedagogical knowledge that is contained within these patterns in a more abstract form that could be used for a variety of purposes. I also refer to this ideal within my thought about a [[Unified theory of pedagogy|unified theory of pedagogy]]. Whilst it may seem abstract, the creation of some unified model could be used by a myriad of different application technologies such as lesson planning or [[Intelligent tutoring system|intelligent tutoring systems]]. ## The tradeoff between expressiveness and utility Modelling languages are inherently reductive in that they take the complexities of the world and reduce it to structured forms. Here, it seems that expressiveness and utility are constantly battling each other, defined as: * **Expressiveness,** is the extent to which the nuance of pedagogical knowledge can be described through it. * **Utility,** is the different capabilities that is the design of the language affords, such as sharing, reuse, adaptability, executability and more. Therefore a PML should not be judged solely on one quality, but rather on how it navigates the tradeoff between expressiveness and utility. On the extremes, we have natural language affording deep capabilities for expressiveness but lacking inherent pedagogical structure, execution or building knowledge in a way that can be interoperable; which is the norm for learning science publications. On the other extreme, we have the pedagogy model of [[Intelligent tutoring system|intelligent tutoring systems]] that prioritise executability at the tradeoff of social and stochastic dimensions to learning. Though this tradeoff is continuously evolving over time, since the utility of a PML changes as societal needs and technological affordances evolve. For example, the rise of [[Large language model|LLMs]] gives newfound affordances to the PML of natural language. Additionally, changing societal values on the [[Purpose of education|purpose of education]] to civic participation and the development of virtues can devalue the utility of the strict pedagogy model of [[Intelligent tutoring system|intelligent tutoring systems]].