Semantic Technologies and Cultural Heritage

Understanding the knowledge generation processes related to cultural heritage

César González-Pérez

16/05/2022

Conceptual modelling Discourse Semantic technologies Transdisciplinarity

Friends and family often ask me what my job is, and it's always very difficult for me to explain, beyond just saying that I'm a researcher. With this post I try to answer this dreaded question. It is true that I am a researcher, and I have been for a long time. As things are better understood through their names, I will say that I do research on semantic technologies and cultural heritage. This is the name of the research line that I launched many years ago, and on which I continue to work today.

This name refers to, on the one hand, semantic technologies. Semantics is, roughly, the relationship between the words we use and the things they refer to. For example, when I say “I find that monument impressive” while pointing to the cathedral of Santiago de Compostela, the semantics is the relationship that exists between “that monument” and the cathedral in front of me. Semantics has traditionally been part of linguistics and, in Spain, part of philology, despite the fact that other sciences also study meaning from different perspectives, such as psychology, philosophy, or computer science. Now, semantic technologies are the applications that we can make of semantics in order to solve practical problems. For example, the automatic translator Google Translate that many of us often use is a semantic technology.

On the other hand, the name of my line of research includes cultural heritage. According to CHARM, cultural heritage is the set of things that represent our cultural values and that, in addition, can provide benefits (not necessarily economic) to us in the future. We usually understand cultural heritage as those things that we want to preserve because they are a legacy from the past and represent, in some way, values, events, or situations that we want to remember and that future generations should remember as well. I have used the word “represent” several times, and this points precisely to the discussion about semantics of the previous paragraph.

When we work with cultural heritage, we realise that it is created, modified and even destroyed by virtue of certain discourses. For example, we all consider that the cathedral of Santiago de Compostela is part of our cultural heritage because that is how it is explained and justified by a number of discourses that we consider valid. In this way, the discourses and, in particular, their semantics, are what supports the processes of cultural heritage construction and conformation. Thus, the line of research on semantic technologies and cultural heritage aims to understand how things become heritage through the discourses, arguments and meanings that are constructed by people, and how these things, once they are cultural heritage, are received, modified and interpreted by different groups through new discourses, arguments and meanings.

For example, imagine that someone says that “it is necessary to keep masses in the Cathedral because we are a culture rooted in Catholicism, and not to do so would be giving up our fundamental values.” We can study this statement from a discursive point of view, looking at what the speaker means, what things in the world they refer to (the Cathedral, but also Catholicism, our culture, and our values), and how they justify what they demand (we shouldn’t give up our values). To do this systematically and scientifically, we need, on the one hand, to represent the world (or small parts of it) in a simplified and rigorous way and, on the other hand, to analyse discourses also in a rigorous and reliable way. To respond to these needs, we have, on the one hand, conceptual modelling, a technique that emerged within software engineering but that today is used to build and manage rigorous representations in many different areas, such as biomedicine or legal sciences. And, on the other hand, we have discourse analysis, a set of techniques that help us break down a discourse into its basic elements and understand how it works in terms of structure, meaning, argumentation and social context.

Evidently, working with cultural heritage in this manner impels us to work with different disciplines (software engineering, philosophy of language, linguistics, heritage studies), which are often considered to be distant and weakly connected to each other. To achieve this, we have designed a strategy of transdisciplinary co-research. By “co-research” we mean that we carry out research in two or more disciplines, we generate new knowledge in all of them, and we publish our results in academic journals and conferences belonging to all of them. By “transdisciplinary” we mean that this simultaneous advance occurs thanks to the fact that the work in each of the disciplines involved not only advances the state of the art in that discipline, but also helps to advance it in the others. For example, between 2018 and 2020 we carried out a project in which we applied the ISO/IEC 24744 standard on software development methodologies to build a collaborative ethnographic methodology. This advanced the state of the art in ethnographic methodologies. At the same time, the experience revealed that there were aspects, such as consensual decision-making or the dynamic modification of the methodology while it is being applied, that were not addressed by ISO/IEC 24744, so we proposed an improvement to the standard, which advances the state of the art in software engineering. It is unlikely that this improvement in the standard would have occurred had it not been applied to a humanistic field. And, similarly, it is unlikely that the world of collaborative ethnographic methodologies would have advanced as it has if we had not applied a standard from the world of software engineering.

This approach, obviously, defeats any disciplinary paradigm, since it is incompatible with the traditional organisation of science around clear and perfectly defined areas or fields of activity. Although Incipit CSIC has always been a problem-oriented centre, and very receptive and efficient when it comes to housing specialists from various fields, the Spanish, European and international science systems are almost never capable of understanding and adequately managing truly transdisciplinary approaches. To mitigate this problem, from the Semantic Technologies and Cultural Heritage line we have developed a network of external collaborators with whom we work closely and continuously.

To begin with, Patricia Martín-Rodilla, lecturer of software engineering at the University of A Coruña, collaborates with us on issues related to conceptual modelling and discursive information extraction. Martín Pereira-Fariña, lecturer of philosophy of language at the University of Santiago de Compostela, and Beatriz Calderón-Cerrato, researcher in linguistics, work with us on issues related to argumentative and discursive analysis. Isabel Cobas-Fernández, archaeologist and secondary school teacher, collaborates with us to study the heritage discourses of children and teenagers. Leticia Tobalina, archaeologist and researcher at Casa de Velázquez, works with us on aspects related to information vagueness in conceptual modelling. Others, such as Alejandro López and Raquel Liceras, are also beginning to work with us on issues related to the uses of heritage discourses. We must also thank many other national and international people for their collaboration, such as Antonio Vallecillo, Brian Henderson-Sellers, Charlotte Hug, Chris Reed, Elea Giménez, Fátima Díez, Jeremy Huggett, Juan Antonio Barceló, Manuel Miguéns, Maria Elena Castiello, Nacho Vidal, Óscar Pastor, Pablo Gamallo, Rebeca Blanco, Ruth Varela or Xabier Larrucea. And, of course, we also thank the colleagues at Incipit, who are always open to collaborations and exchanges between different disciplines.

At the same time, we have always wanted our work to be of a broad spectrum. In other words, almost everything we do encompasses theoretical research, methodological research, technological development, and the construction of tools based on the above. In this way, the tools we produce have solid theoretical and methodological support, and the theories we develop are not mere paper-based speculations but, when implemented in a concrete way, are tested and validated in real-world situations. Over the years, we have developed technologies and tools such as ConML, Bundt, KaleidoScapes, Cabila, IAT/ML or [link:http://www.iatml.org/">LogosLink, all of them publicly available (or will be soon), and all of them supported by theoretical developments of our own or in close collaboration.



To summarise, I will finish by saying that the research line on Semantic Technologies and Cultural Heritage at Incipit CSIC seeks to understand and assist the processes of knowledge generation and modification that occur in relation to cultural heritage through its representation and the analysis of the discourses that convey it.

Semantic Technologies and Cultural Heritage

Understanding the knowledge generation processes related to cultural heritage

César González-Pérez
16/05/2022
Conceptual modelling Discourse Semantic technologies Transdisciplinarity

Friends and family often ask me what my job is, and it's always very difficult for me to explain, beyond just saying that I'm a researcher. With this post I try to answer this dreaded question. It is true that I am a researcher, and I have been for a long time. As things are better understood through their names, I will say that I do research on semantic technologies and cultural heritage. This is the name of the research line that I launched many years ago, and on which I continue to work today.

This name refers to, on the one hand, semantic technologies. Semantics is, roughly, the relationship between the words we use and the things they refer to. For example, when I say “I find that monument impressive” while pointing to the cathedral of Santiago de Compostela, the semantics is the relationship that exists between “that monument” and the cathedral in front of me. Semantics has traditionally been part of linguistics and, in Spain, part of philology, despite the fact that other sciences also study meaning from different perspectives, such as psychology, philosophy, or computer science. Now, semantic technologies are the applications that we can make of semantics in order to solve practical problems. For example, the automatic translator Google Translate that many of us often use is a semantic technology.

On the other hand, the name of my line of research includes cultural heritage. According to CHARM, cultural heritage is the set of things that represent our cultural values and that, in addition, can provide benefits (not necessarily economic) to us in the future. We usually understand cultural heritage as those things that we want to preserve because they are a legacy from the past and represent, in some way, values, events, or situations that we want to remember and that future generations should remember as well. I have used the word “represent” several times, and this points precisely to the discussion about semantics of the previous paragraph.

When we work with cultural heritage, we realise that it is created, modified and even destroyed by virtue of certain discourses. For example, we all consider that the cathedral of Santiago de Compostela is part of our cultural heritage because that is how it is explained and justified by a number of discourses that we consider valid. In this way, the discourses and, in particular, their semantics, are what supports the processes of cultural heritage construction and conformation. Thus, the line of research on semantic technologies and cultural heritage aims to understand how things become heritage through the discourses, arguments and meanings that are constructed by people, and how these things, once they are cultural heritage, are received, modified and interpreted by different groups through new discourses, arguments and meanings.

For example, imagine that someone says that “it is necessary to keep masses in the Cathedral because we are a culture rooted in Catholicism, and not to do so would be giving up our fundamental values.” We can study this statement from a discursive point of view, looking at what the speaker means, what things in the world they refer to (the Cathedral, but also Catholicism, our culture, and our values), and how they justify what they demand (we shouldn’t give up our values). To do this systematically and scientifically, we need, on the one hand, to represent the world (or small parts of it) in a simplified and rigorous way and, on the other hand, to analyse discourses also in a rigorous and reliable way. To respond to these needs, we have, on the one hand, conceptual modelling, a technique that emerged within software engineering but that today is used to build and manage rigorous representations in many different areas, such as biomedicine or legal sciences. And, on the other hand, we have discourse analysis, a set of techniques that help us break down a discourse into its basic elements and understand how it works in terms of structure, meaning, argumentation and social context.

Evidently, working with cultural heritage in this manner impels us to work with different disciplines (software engineering, philosophy of language, linguistics, heritage studies), which are often considered to be distant and weakly connected to each other. To achieve this, we have designed a strategy of transdisciplinary co-research. By “co-research” we mean that we carry out research in two or more disciplines, we generate new knowledge in all of them, and we publish our results in academic journals and conferences belonging to all of them. By “transdisciplinary” we mean that this simultaneous advance occurs thanks to the fact that the work in each of the disciplines involved not only advances the state of the art in that discipline, but also helps to advance it in the others. For example, between 2018 and 2020 we carried out a project in which we applied the ISO/IEC 24744 standard on software development methodologies to build a collaborative ethnographic methodology. This advanced the state of the art in ethnographic methodologies. At the same time, the experience revealed that there were aspects, such as consensual decision-making or the dynamic modification of the methodology while it is being applied, that were not addressed by ISO/IEC 24744, so we proposed an improvement to the standard, which advances the state of the art in software engineering. It is unlikely that this improvement in the standard would have occurred had it not been applied to a humanistic field. And, similarly, it is unlikely that the world of collaborative ethnographic methodologies would have advanced as it has if we had not applied a standard from the world of software engineering.

This approach, obviously, defeats any disciplinary paradigm, since it is incompatible with the traditional organisation of science around clear and perfectly defined areas or fields of activity. Although Incipit CSIC has always been a problem-oriented centre, and very receptive and efficient when it comes to housing specialists from various fields, the Spanish, European and international science systems are almost never capable of understanding and adequately managing truly transdisciplinary approaches. To mitigate this problem, from the Semantic Technologies and Cultural Heritage line we have developed a network of external collaborators with whom we work closely and continuously.

To begin with, Patricia Martín-Rodilla, lecturer of software engineering at the University of A Coruña, collaborates with us on issues related to conceptual modelling and discursive information extraction. Martín Pereira-Fariña, lecturer of philosophy of language at the University of Santiago de Compostela, and Beatriz Calderón-Cerrato, researcher in linguistics, work with us on issues related to argumentative and discursive analysis. Isabel Cobas-Fernández, archaeologist and secondary school teacher, collaborates with us to study the heritage discourses of children and teenagers. Leticia Tobalina, archaeologist and researcher at Casa de Velázquez, works with us on aspects related to information vagueness in conceptual modelling. Others, such as Alejandro López and Raquel Liceras, are also beginning to work with us on issues related to the uses of heritage discourses. We must also thank many other national and international people for their collaboration, such as Antonio Vallecillo, Brian Henderson-Sellers, Charlotte Hug, Chris Reed, Elea Giménez, Fátima Díez, Jeremy Huggett, Juan Antonio Barceló, Manuel Miguéns, Maria Elena Castiello, Nacho Vidal, Óscar Pastor, Pablo Gamallo, Rebeca Blanco, Ruth Varela or Xabier Larrucea. And, of course, we also thank the colleagues at Incipit, who are always open to collaborations and exchanges between different disciplines.

At the same time, we have always wanted our work to be of a broad spectrum. In other words, almost everything we do encompasses theoretical research, methodological research, technological development, and the construction of tools based on the above. In this way, the tools we produce have solid theoretical and methodological support, and the theories we develop are not mere paper-based speculations but, when implemented in a concrete way, are tested and validated in real-world situations. Over the years, we have developed technologies and tools such as ConML, Bundt, KaleidoScapes, Cabila, IAT/ML or [link:http://www.iatml.org/">LogosLink, all of them publicly available (or will be soon), and all of them supported by theoretical developments of our own or in close collaboration.



To summarise, I will finish by saying that the research line on Semantic Technologies and Cultural Heritage at Incipit CSIC seeks to understand and assist the processes of knowledge generation and modification that occur in relation to cultural heritage through its representation and the analysis of the discourses that convey it.