The deep and long running influence of John Unsworth's notion of "scholarly primitives" is undeniable. Around the same years W. McCarty and H. Short developed the idea of the “methodological commons”, another framework used to map the kind of scholarly activity in the field that was then called Humanities Computing. In many ways it was better at describing the field, but “scholarly primitives” were more successful and had a deeper impact in the self-conceptualization and meta discourse of DH. One may wonder why the idea proposed by Unsworth was so successful. I think that the success of Unsworth's scholarly primitives is rooted in the extreme simplicity of their definition, or better, in the pragmatist framework that governed their formulation. Apart a (somewhat instrumentalist) hint to the notion of primitive terms in axiomatic theories as an analogical base for its proposal, both the selection and the description of the scholarly primitives in Unsworth's paper were based on practical examples and anecdotal exemplification. After all, they were developed in the preparation of two project proposals. This pragmatist approach was of course very easy to grab for the community, in as much as it made appeal to the personal experiences of many practitioners of DH. But there is also a deeper reason: it was ideologically and culturally consonant with the phase from the evo of Humanities Computing to that of Digital Humanities. In a sense, it served as base for the rhetoric of scholarly inclusivity that was going to characterize the field and that some years was epitomized by the notorious “Big Tent DH” motto adopted in DH2011 conference at Stanford.

The primitives listed by Unsworth were so generic and encompassing that could be used to justify the inclusion inside the field of almost everyone had to with digital resources and/or computational tools, at whatever level of methodological and theoretical involvement. They were a perfect justification for accepting as a valid and sound research in DH both “hard core” stylometry, markup practice, ontology building and “simple” Web sites development. One aspect worth pointing out is that most of the primitives were activities not inherent to the research and new knowledge production activity, but to its infrastructural pre- and post-conditions.

Although I don't want to overlook their historical importance, I think it is necessary to rethink the methodological foundations of the galaxy shaped field of DH in a more epistemologically sound way. My proposal is based on three core layers:

1) the computational methods layer, that pertains to the set of formal methods adopted in research

2) the data model layer, that pertains to the set of data models and data structures used in research

3) the theory/model layer, that pertains to the way theories and models control the research process and gives

Each layer on its part is structured in typologies. For instance, methods are divided in two main sets: fully automatic computing methods vs. human based computing methods. Real and innovative DH research is composed by an interplay of all three layers. The critical (in a Kantian sense) and theoretical study of this methodological framework and of its interaction with humanities theories and scholarship forms the subject of a real disciplinary core of Digital Humanities studies, for which maybe the label Humanities Computing should be taken back to intellectual life.