Chapter 16

The Configurational Building Blocks of Models of Description

 

16.1. Introduction

In the previous chapters, we have presented the meta-lexicon of conceptual terms* in the ASW universe of discourse* and the thesaurus which provides a whole range of pre-defined values for interpreting a particular conceptual term or configuration of conceptual terms.

In this chapter, we shall discuss how the conceptual terms and the thesaurus integrate with one another to form a model of description or a library of models of description particular to a given audiovisual archive.

In section 16.2, we shall recap the concept of a model of description – a concept which we have used throughout this book.

Sections 16.3 and 16.4 are given over to a presentation of what exactly a sequence of description is. We shall also see that every sequence of description has a place in a library of sequences of description. Two important points which we shall discuss in these two sections are, firstly, the question of defining a sequence of description using a canonic pair of schemas of definition* and, secondly, that of selecting and introducing a sequence into a model of description*.

Finally, section 16.5 will again present the level of schemas of definition for a descriptive model – schemas of definition which, as we know, are divided into two major categories: schemas of definition of the objects of analysis* and schemas of definition of the activities of analysis*.

16.2. Analysis of an audiovisual text and models of description

As has already been mentioned, a model of description* of an audiovisual text or corpus is made up of at least one, but usually several, local sequences*. Each sequence, in turn, is made up of at least one schema defining the object of analysis* (of the descriptive sequence) and one schema defining the procedure of analysis*.

The schema defining the object of analysis is specified by at least one conceptual term denoting the type of objects or domains belonging to the ASW universe of discourse*. The schema specifying the procedure of analysis* is made up of at least one descriptive activity* in the ASW universe of discourse according to which the object of analysis* should be indexed (in the broad sense of the term; see Chapters 1 and 14).

We can clearly see that a model of description has an organization which is simultaneously hierarchical, modular and configurational:

– a hierarchical organization in the sense that it is made up of more local units which are, in particular, the sequences and schemas;

– a modular organization in the sense that each sequence and schema forms an autonomous structural unit which may belong to various models of description (for the sequences) or to various sequences (for the schemas of definition);

– a configurational organization in the sense that a conceptual term denoting either a type of object* to be analyzed (and therefore stemming from the metalexicon of conceptual terms beginning with the root term [Object of analysis]) or an activity of analysis* (and therefore belonging to the meta-lexicon of conceptual terms beginning with the root term [Procedure of analysis]) can only integrate a model of description as an entity positioned in relation to all the other entities selected by the model via its schemas of definition (of analytical objects and procedures) and sequences of description.

This very particular, sophisticated and restrictive organization of a model of description relates directly to our conception of analysis* of an audiovisual text or corpus as outlined in Chapter4.

Remember that we distinguished between four main levels of analysis (see Figure 4.3; section 4.4). The first and most general level is that of choosing a certain type of analysis. In this book, we have favored the analysis of the audiovisual content, which is one type among many. The second level is determined by the choice made at the first level (i.e. by the choice of the type of analysis). This second level identifies the different analytical tasks to be carried out in order to perform a chosen type of analysis. For instance, as Figure 4.3 shows, analyzing audiovisual content includes, amongst other tasks, referential description* or description of the discourse production* around an object thematized in an audiovisual text or corpus.

The third level identifies the analytical procedures which are appropriate for carrying out an analytical task. In this book, we have given priority to two very common procedures which can be used in a great many analytical tasks – the procedures of free description* and controlled description*. Finally, the fourth level identifies the specific analytical activities which define a procedure of analysis. These activities, as we know, are all defined in the meta-lexicon of conceptual terms beginning with the root term [Procedure of analysis].

This four-level vision of the activity of analysis is in fact directly expressed in the hierarchical, modular and configurational organization of an ASW model of description.

Thus, a specific type of analysis (1st level, Figure 4.3) manifests itself in the form of one or more models of description. For instance, analysis of the content of an audiovisual text, which is a specific type of analysis, requires one or more models of thematic description* which belong to a library of models of description* – a library which we use to analyze, for example, the objects* belonging to the universe of discourse* of an audiovisual archive.

A specific analytical task (2nd level, Figure 4.3) manifests itself in the form of one or more sequences of description. As we know, a sequence of description forms a specific building block* used to create a model of description. Thus, the task of referential description*, aimed at identifying and explicitizing the domain or the knowledge object thematized in an audiovisual text, is supported by one or more sequences which are functionally specialized for that task (see our example in Chapter3).

A procedure of analysis (3rd level, Figure 4.3) in turn is supported by one or more schemas of definition*. As explained in the previous chapter, the schema of definition of a descriptive activity *, along with the object of the analysis (i.e. the object to be described), forms a sequence of description. For instance, the schema of the procedure of free description* can make up not only a sequence which is functionally specialized in the referential description of a domain of knowledge, but also one which is specialized in the description of the discourse production around the domain of knowledge thematized, a sequence that is functionally specialized in the description of the audiovisual or verbal expression of the thematized domain of knowledge, etc. (see our examples in Chapter3).

Finally, an analytical activity (4th level, Figure 4.3) is defined in the meta-lexicon of conceptual terms* denoting all analytical activities in the ASW universe of discourse. One or more descriptive activities make up a schema of definition. Thus, the activity [Drafting a summary presentation] may – along with other descriptive activities – define the schema of free description*, the schema of controlled description*, the schema of creating a meta-textual commentary, the schema of iconographic illustration, etc.

A particularly important role here belongs to two main types of configurations (or modules) which not only make up all models of description but also define the specific relationships maintained between the conceptual terms* selected by or in a model of description. These two types of configuration are:

– the sequences of description and;

– the schemas defining both the objects and activities of analysis.

16.3. The library of sequences making up the model of thematic description

We limit ourselves here to the sequences which make up the models of thematic description*, i.e. the models which serve to describe the content of an audiovisual text or corpus. As we know, models of thematic description form one particular type of descriptive model* amongst many others, including, for example, models of audiovisual description* stricto sensu (that is, description of the visual and/or sound shots), models of pragmatic description* (adaptation of a text to the expectations and skills of a target audience) or indeed models of paratextual description* (description of the formal identity of an audiovisual text).

A model of description is made up of one or several functionally specialized fields of analysis. Each functional field of a model of description consists of one or more local structures called sequences. A sequence serves for describing a specific part or aspect of the object to be described. Every sequence making up a model of description is thus thematically and functionally specified in relation to a specific field of analysis. In particular, the following five can be identified:

– the field of referential description of the knowledge object(s) thematized in an audiovisual text or corpus. Here we identify and describe, for instance, and returning to the experimentation workshops in the ASW-HSS project, a particular culture, work of literature, archaeological dig, scientific discipline, scientific personality, and so on;

– the field of referential contextualization of the knowledge object(s) thematized in the audiovisual text/corpus. This field can be broken down into a field of description called Spatial and/or geographic location of the domain of knowledge, a field of description called Temporal and historical location of the domain of knowledge and sometimes a field called Thematic contextualization (i.e. institutional, social, epistemological…) of the domain of knowledge;

– the field of analysis of discourse production around the thematized knowledge objects. This field of analysis enables the analyst to provide information about the particularities of the enunciative and discursive treatment of a given subject in a video. Who is the person responsible (the enunciator) for the thematization of the object or domain of knowledge? From which point of view is the topic approached? What is the discursive hierarchy of the information (that is, which pieces of information are the most important, the most central, and which pieces of information are less so? What is the level of specialization in the treatment of the thematized object? Which information is assumed to be known and which pieces of information are treated as novel? These and many more such questions are posed when examining discourse production;

– the field of analysis of the verbal expression (not developed in the context of our research) and the audiovisual expression of the knowledge objects thematized and turned into discourse. This field enables the analyst to produce information about the significant particularities of the verbal language (but also gestural, postural, somatic) and audiovisual language used to express and communicate the discourse about an object or domain of knowledge;

–the field of meta-textual analysis of the commentary. This field offers the analyst the option of more explicitly expressing his point of view either about his own analysis or about the way in which the domain is dealt with in the source video.

Above all, let us take from this that a field of analysis always has a corresponding library of functionally appropriate sequences. Figure 16.1 shows an extract from such a library of sequences that we use, in our case, to define the library of models for describing the content of the audiovisual texts making up the collection of the FMSH-ARA audiovisual archives¹ devoted to the production and diffusion of scientific events at the Fondation Maison des Sciences de l’Homme in Paris.

As we can see in Figure 16.1, the library of sequences is made up of 5 major collections – namely:

– A collection of sequences entitled Analysis of the domain of expertise, shown in greater detail in Figure 16.2. This collection encompasses all the sequences that are functionally specialized in the identification and explicitation (description, indexation, etc.) of the objects and domains of knowledge thematizable in the audiovisual production of the FMSH-ARA archives. In other words, this collection constitutes the necessary resources to define the first functional field of a model of thematic description (of the audiovisual content), i.e. that reserved for the task of referential description (in the context of analysis of the audiovisual content which – let us stress once more – forms a particular type of analyses of the audiovisual text or corpus). For the moment, the collection of sequences in question specifies referential subjects such as Scientific research in HSS, Visions, ideologies and representations, Social identities or Collective memories. The collection of sequences representing these specialized referential fields of analysis can, at any moment, be supplemented by new sequences. Given that the universe of discourse* of an archive is not set in stone, everything depends here on the specificity of the corpus or corpora at stake and on the policy of diffusion and valorization of scientific heritage (i.e. the answer to questions like: “Which subjects, which geopolitical regions, which eras, etc. should be prioritized, for what audience and what use context?”).

– A collection of sequences entitled Location of the domain of expertise (see Figure 16.1). Figure 16.3 shows its organization in the context of defining the metalanguage of description* of the universe of discourse* of the audiovisual collection making up the FMSH-ARA archives. As we can see, this collection is divided into one set of sequences used for the spatial (geographic or geopolitical) pinpointing of an object or domain of knowledge thematized in a text, and a second set of sequences used to locate it temporally (chronologically). As we have already said above, there is nothing standing in the way of this collection being further specified in order to be able to better pinpoint an object or domain of knowledge dealt with in an audiovisual text.

– A collection of sequences entitled Analysis of the discourse production around the subject (Figure 16.1). This collection was not extensively developed when developing the metalanguage of description of the audiovisual texts belonging to the FMSH-ARA archives. It contains only one set of sequences which we defined in order to better identify certain acts of discourse used to “speak” about an object or domain of knowledge (see Figure 16.1). In our example, it is primarily a question of enabling the analyst to identify the acts of definition and exemplification (of this-orthat notion, this-or-that object, such-and-such an expression, etc.) in an audiovisual text or corpus. The results of such an analysis can be used to put specialized interfaces in place on the Web portal of the FMSH-ARA archives, for accessing audiovisual passages in which the interested audience will find definitions and concrete examples pertaining to a given issue.

– A collection of sequences entitled Analysis of the audiovisual expression of the subject (Figure 16.1). As this figure shows, this collection is only developed for analyzing the audiovisual mise en scène of an object or domain dealt with in a text: visual point of view, framing and visual shots, soundscape, etc.

– Finally, a collection of sequences reserved for the analyst’s meta-discourse – a collection which (in our example) contains only one sequence, enabling the analyst to add free commentary to his description of such-and-such a subject.

– Every library of sequences of description is organized in a manner identical to the example briefly presented above. However, it should not be forgotten that in this example, we have only developed those collections of sequences which we need in order to specify the descriptive models enabling us to carry out a very specific type of analysis – the description of the audiovisual content* (see Figure 4.3, which identifies the main descriptive tasks which must be carried out for this type of analysis). For the library of sequences to be complete, we would have to add the sequences and collections of sequences needed in order to carry out the tasks of the other main types of analysis – namely paratextual description*, pragmatic description* and indeed audiovisual description* stricto sensu.

Any sequence can be re-used in different libraries of descriptive models. In other words, the same sequence can serve to define the metalanguage of description of the universes of discourse of different audiovisual archives. For instance, the sequence Analysis “Scientific research in HSS” is a sequence used not only for describing the corpora belonging to the FMSH-ARA archives but also for describing corpora from other archives for which the domain of scientific research in HSS is pertinent.

This point leads directly on to the importance of clearly distinguishing between the different levels of integration which characterize the metalanguage of description* of the universe of discourse* of an audiovisual archive. Thus, we must distinguish between:

– the level of the conceptual terms and the level of the schemas of indexation for selecting and informing a conceptual term (or configuration of conceptual terms) (see Chapter14);

– the level of the schemas of definition, which can be broken down into schemas of definition of the analytical objects and schemas of definition of the analytical activities (see below);

–the level of sequences of analysis and that of collections of sequences of analysis;

– the level of models of description (thematic, audiovisual, paratextual, metatextual, etc.);

– and finally, the level of the collection of models of description and of the library* stricto sensu of models defining the metalanguage of description* of the universe of discourse* of a given audiovisual archive.

16.4. Definition and insertion of a sequence into a model of description

Every sequence forms a structural and functional object with its own internal organization. Let us take a brief look at the example of the sequence Analysis “Scientific research in HSS” (Figure 16.4).

As Figure 16.4 shows, a sequence can be deployed in several, more specialized sub-sequences which occur in accordance with an explicit syntagmatic order.

In our example, the syntagmatic relation is that of unilateral presupposition which we have already encountered many times. This requires the analyst to first indicate the context of the object or domain thematized in the audiovisual text. In our case, it is a question of specifying the scientific disciplines spoken about in an audiovisual text (see, in Figure 16.4, the sequence entitled Analysis of the theme “Discipline”). Once this part of the subject has been indicated, the analyst may, in a second stage, specify certain points of the thematized object or domain. In our case, the analyst is offered the chance to produce information – if available – about the research activities, the domain(s) of research or indeed the theme(s) of research (see, in Figure 16.4, the sequences entitled Analysis of the theme “Research activity”; Analysis of the theme “Research domain”; Analysis of the theme “Research topic”).

Let us take from our brief discussion of the example of referential description of the domain, research in human and social sciences, that in its simplest internal organization, a sequence presents as a single entity, made up, as we shall very shortly see, of at least two schemas of definition – a schema of definition of the object of analysis* and a schema of definition of the activity of analysis*. However, it can also be deployed as a whole network of sub-sequences, positioned in relation to one another according to a precise syntagmatic structure.

Finally, a syntagmatic organization of sequences such as that shown in our example (Figure 16.4) may be supplemented by new sub-sequences which constitute either alternatives to a sub-sequence already identified and defined in the existing syntagmatic organization or complements to it, additions.

Let us take the example of the sub-sequence Description of the theme “Research activity”. This is defined and positioned as a simple sequence, as an indivisible entity in the configuration of sequences which we use to analyze subjects relating to research in human and social sciences (Figure 16.4). However, there is nothing standing in the way of alternative sequences to this one being defined – alternative sequences which enable us to adapt the task of analysis to the particularities of a given audiovisual corpus or archive. When the macro-sequence entitled Analysis “Scientific research in HSS” is selected to figure in a model of thematic description, then the sub-sequence Description of the theme “Research activity” will not be selected as is shown in Figure 16.4, but rather, one or other of its alternatives. These remarks also apply to a case where one or more specialized sequences have to be added to an existing configuration of sequences.

Every sequence – be it simple or composed of two or several more specialized sequences – is a relational entity which, on the one hand, positions the conceptual schemas of definition* in relation to one another and which, on the other hand, is positioned in relation to other sequences to form a model of description*.

As has already been mentioned on many occasions, every sequence is defined by at least two if not several conceptual schemas of definition. Figure 16.5 shows how a given sequence of description is defined, in our case via the interface of the OntoEditor tool in the ASW Modeling Workshop, by identifying the appropriate canonic pair of schemas of definition.

In our case, the sequence First: Description of the main theme (which is part of the macro-sequence Analysis “Intercultural dynamic”) is thus defined:

– by a conceptual schema specifying the appropriate analytical objects (see Figure 16.5, on the right, which shows the shortcut to the schema Choice of the appropriate CT(s)); and

– by a conceptual schema specifying the procedure of description which, here, is free description* in its simplified form (that is, the analyst only has to fill in one field of indexation – that of the minimal expression; see Chapter14 for further information).

Figure 16.6 illustrates the process of integration of a sequence (in our case, a macro-sequence) into a specific model of thematic description. In our case, the macro-sequence Analysis “Intercultural dynamic” is integrated with four other macro-sequences to form model of content description Subject “Intercultural dynamics”:

– the first macro-sequence in question here serves to explicitize (describing, interpreting) a fact considered (by the analyst, the author of the analyzed text, etc.) as representing a concrete case of the intercultural dynamic;

– two macro-sequences serve to pinpoint the fact being analyzed (the macrosequence Location in a region or country of the world and the macro-sequence Temporal location by period and/or precise date)

– and finally, one macro-sequence serves to describe the discourse production around the subject (the macro-sequence Analysis of the rhetorical acts used to speak about the subject).

16.5. Summary presentation of a library of schemas of definition

As we know, a sequence is made up of at least two schemas of definition, one of which must belong to the category of schemas Analytical Object and the other to the category Analytical procedure

Each archive has its own library of schemas. Thus, Figure 16.7 shows an extract from the library of schemas of definition for the FMSH-ARA archives. This is made up of two major collections corresponding to the two main categories of schemas previously cited.

The internal organization of the schemas into collections of more specialized schemas corresponds, in terms of the schemas of definition of the analytical objects, to the different types of objects* pertinent for the description of an audiovisual text or corpus:

– referential objects (i.e. which belong to such-and-such an empirical domain documented by an audiovisual corpus);

– contextual objects (for spatial/temporal location, etc.);

– discursive objects (i.e. pertaining to discourse production);

– verbal and audiovisual objects (i.e. relating to visual expression, acoustic expression, syncretic expression, etc.);

– reflexive objects (i.e. which serve to produce a metadiscourse about the analysis itself).

A library of schemas of definition contains all the schemas involved in constituting the models of description*. However, as a general rule, it is endowed with many more schemas of definition which may, potentially be relevant for elaborating new models of description or indeed models of description, “alternative” in relation to those which exist and belong to a library of models of description* defining the universe of discourse* of an audiovisual archive.

Let us now take a look at the internal organization of a schema of definition. As has already been mentioned many times, every schema of definition is made up of one or more conceptual terms* which stem either from the meta-lexicon of conceptual terms denoting analytical objects in the ASW universe of discourse*, or from the meta-lexicon of conceptual terms denoting the analytical activities.

The simplest schema of definition is characterized by the selection of one and only one conceptual term. This selection equates to the affirmation of a unary relation of identification of a conceptual term. Figure 16.8 shows an example of this. The schema defining the analytical object Social movement relies solely on the conceptual term [Social movement].

However, a schema can also select several conceptual terms whether or not they belong to the same taxonomic domain of the meta-lexicon of conceptual terms. An example is shown in Figure 16.9. This schema defines an analytical object entitled Citizenship, social inclusion and exclusion. In order to do so, this schema draws upon a set of conceptual terms such as [Status of citizen], [Status of foreigner], [Status of stateless person], etc., from which the analyst can choose the most appropriate term(s) to index an audiovisual text which deals with the question of citizenship and/or social inclusion/exclusion.

A third and final scenario is that of the “grand schema of definition (of an object of analysis* or procedure of analysis*)” or, as we prefer to call it, the macro-schema of definition (in the same way as a macro-sequence). A macro-schema is made up of various more specialized schemas which are positioned in relation to one another. Each more specialized schema includes at least one conceptual term (but, as the example in Figure 16.9 shows, a schema can also include various conceptual terms).

Figure 16.10 shows the breakdown of the schema of definition of the analytical object Socio-historical processes of the modern world into a whole series of more specialized schemas: the schema Globalization, the schema Urbanization, the schema Migration, etc.

The macro-schema Socio-historical processes of the modern world can be selected just as it is in a sequence of referential description. In this case, it offers the analyst a whole series of options for configuring the structure of his topic (see Chapter 5 for further explanations). It is also possible to select only certain schemas from a macro-schema of definition such as that shown in Figure 16.9.

---

1 See http://semiolive.ext.msh-paris.fr/ASW-HSS/.