The philosophical twentieth century was characterized by an explosion of science and scientific nature which highlighted an idea of progress and progression of the human thought and evolution as something which does not need further questions. The idea that functionality and applicability of a theory can be applied to praxis, which executes and finalizes such theory, comes full circle.
For some years, however, the reflection has started to crumble some principles that so far were thought to pertain exclusively to the scientific domain, demonstrating that among engineering problems there is also a certain need for a theoretical reflection that has nothing to do with immediate functionality.
Starting from artificial intelligence, we have tried to show the need of an ontological thought and its rediscovery in recent years in the context of the studies of landscape and its geomorphology.
Philosophy survives and feeds on questions. The many questions posed by the philosophical tradition could be distinguished into two macro-typologies: questions born away from the world of theoretical speculation, to which we want to give a conceptual answer, that is, that which pertains to knowledge, the moral, epistemological questions; and questions aimed at speculation for its own sake, sometimes of self-referential nature.
Every question is legitimate, especially in philosophy where this law is even more valid, so much so that every question, even the ones belonging to the first typology, needs to be asked in relation to a broader field and by clarifying better its nature.
In this way, maybe a bit muddled and circular, we wanted to introduce an emerging distinction in the philosophical speculation aimed at seeking knowledge: the distinction between one question which could be defined metaphysical and an ontological question. As it emerges from the essay by one of the major contemporary philosophers, Kevin Mulligan, Truth and ontology, the role of the ontological question is to offer a necessary question, whose answer does not need other relations, nor other questions.
To start with the philosophic tradition, in his essay Mulligan claims that despite the nineteenth century and the philosophical currents are often advocate of a death of ontology and metaphysic, the need for a more general question has remained innate in the philosophical tradition even when it shifted its focus on other reasons for interest. In this way, even the worst enemies of metaphysic, like the philosophers of the Vienna circle, Carnap in the first place, or the followers of Wittgenstein, who made the principle “you can’t talk about things you do not know” a priority, create their vision of the world on the basis of an ontological structure. Before I talk about the ontological rediscovery applied to scientific disciplines it is necessary to ask ourselves and better clarify: what is ontology, then?
If we started from a classical definition, we should say that it is the matter that asks questions about the being as such, that wonders about pure concepts: extension, existence, being. From this Parmenidian and a bit too restrictive definition for our approach, and for a more general approach, we should provide another one with wider ambitions.
We could say: ontology is a purely philosophical way of thinking and reflecting about what surrounds us, by means of a categorical analysis of reality, objects, processes which constitute it. Such interpretation allows to exclude form our definition all the ontologies based on poetics or suggestions, as Mullihan suggests: “ontology cannot be fundamental in any other way than being formal”, that is maintaining a rigorous character.
It is not just precision that makes the structure of ontological reasoning, since ontology as methodology is a need that regulates the stability of our discourses and provides a shared territory to understand the world surrounding us: things, people, living beings, natural elements.
Let’s make a new example. In his essay, Kevin Mulligan asks two series of questions, the first ones metaphysical, the second ones ontological.
Among the first are: is time relational? Are we free? Do values depend on us? The ontological ones are: what is relation? What is existing? What is a substance? From here, it is possible to understand the distinctions and the advancements made by the individual willing to reflect on and question the ontological status of reality. It is possible to notice the rigour and the wide spectrum of issues required to formulate an ontological reflection. It is no coincidence that throughout the years ontology is always more active as an underwriter of postulates which empiric sciences assume (subjects and research fields which are very far away from the apparent self-referentiality of ontological questions, like for example the physical Ontology and the Information science Ontology.
We hope the premise we have just made is exhaustive enough to introduce the complexity of the topic addressed. However, it is necessary to examine more thoroughly the use and rediscovery of ontology. An example is Artificial Intelligence, that is the macrocosm of theoretical/experimental studies that deals with the creation of machines able to carry out, almost autonomously, some operations typical of the human brain. To give a general background of the discipline we quote two emblematic researchers. The first in chronological order is Blaisè Pascal, philosopher and mathematician, who invented the so-called Pascaline, an arithmetic machine, able to do sums and subtractions. It was 1642.
The second of the two emblematic cases is undoubtedly Alan Turing, father and developer of the discipline. Max Newman drewon Turing’s 1936 studies on the so-called Turing machine, to create Colossus, the calculator invented in 1942 to decode the nazi codes codified by the Tunny. Since he was not given credit for it, Turing took his own life biting an apple whose symbology in the future would have become even more iconic than the past one).
Let’s go back on topic, what is asked to artificial intelligence? What does it study? Why does it have anything to do with philosophy or ontology? At first, the study of ontology seems to be extremely distant from the engineering of machines, gap which becomes even bigger if analyzed though the following prejudices: on the one hand, self-referentiality and philosophical speculation, on the other hand, the assumption that technological progress is based on the relation between experimentation and theorization.
According to the words of Minsky Artificial intelligence fulfills the following goals: “making machines do things which would require intelligence if they were done by men”. Following this hint it emerges the need of a biggest distinction and a specification within the development of ontology and in the distinction within philosophical disciplines.
The programming of Artificial intelligence is based in logics, but the contribution given by ontology has a wider range of research. If logics directs its contribution to the study of reasoning, ontology offers instruments of study and analysis to understand the nature of the world and the things around us. Simply put, it would be useful to exemplify this relationship through the philosopher who best highlights the strength of ontological reasoning: Edmund Husserl, a central personality in the eight-nineteenth philosophical century. Husserl is first of all a mathematician, whose rigour is always present in his reasoning. He locates himself in a central position between science and theory. We can focus on the transcendental matter and the objectivity of the perception, without caring too much about a series of perceptual issues which Hussel deals with in his studies
Witough further lingering about the minutia of the philosophy of Husserl: what distinguishes Edmund Husserl from other philosophers is his need to motivate the categorical outside of the subject. In other words, if for Emmanuel Kant and his followers the transcendental is that group of categories with which it is possible to read the outside world, which is and stays something formless, a whole reasonable and shared, for Husserl the transcendental has two poles regulating the perception: the material and the subject. If in Kant the transcendental coincides with subjectivity, in Husserl it resides in the correlation of subject and object, that is in the prior condition of possibility of the experience. In Husserl the transcendental has an objective and subjective implication, the task of the phenomenology (or of the philosopher) is to find laws essential to the experience and to comprehend how the two sides can be combined. The philosopher then studies both the structure of conscience and its intentional acts (subjectivity) and the laws of the so-called “material ontologies”, that is, the objective structures of experience. This essential connection is based on the prior laws which share object and subject, and this creates the ontological reasoning in Husserl and allows us to retrace our steps. The conscience of the subject sees the world with the same logic laws through which the world gives itself to the subject.
Keeping in mind this ontological explanation of sharing the world, we can better understand the need for ontology and its rhythm through artificial intelligence. As Carlo Scognamiglio puts it in his article: “Right now the development of researches on the application of ontology to knowledge engineering poses the exigency to distinguish ontology as technology, that is, an instrument to design databases, from the ontology of the philosophical tradition, which is a categorical analysis, and it reveals to be close to the problems arising from the study of ontology as technology.
With this distinction Scognamiglio clarifies the distinction between ontology as technology and philosophic ontology, which have in common the research of stability of processes and the sharing between the individuals and the community.
In the following pages, the author explains how the study and the design of information systems was based , on the one hand, on a functional conception of knowledge , that is the mechanism of theorization/experimentation. On the other hand, confirming the importance of the return of ontology, today artificial intelligence studies not only focus on the limited object of study, but also look for an ontological approach able to create a common setting, where the database can communicate starting from a common formal language.
Here comes again a problem that ontology has always faced: the problem of ontology as a possibility to establish the stability of common discourses and perceptions. This is an object-reality-language problem, which was present in Plato’s Parmenides and in his criticism of the sophists. It is a problem of stability apriori that cannot be attributed to ideas, that is restrictive to attribute to the subject, but that as we saw in Husserl, and as its continue application to knowledge engineering shows, needs to be a system able to comprehend, in the same logical structure, both the reasoning of the subject and the object, and its way of sitting involuntary in a wider relationships system.
Ontology becomes then a way of reasoning and a common ground upon which comprehension and sharing of the world is based.
In this sense it is easy to understand both the duplicity of ontology and its recent importance: ontology as technology and ontological philosophy play a central role in fitting different worlds within one domain. Quoting again Scognamiglio: “In practice, ontology becomes the guiding principle for the construction of a meta-model which includes and connects different databases, yet contributing to the construction of a single database of a single disciplinary field”.
Bouncing between subject, object, reality, language and representation is a problem that afflicts science and its relationship with our perception of the world. If up to today we have been thinking about artificial intelligence as something interesting but maybe relegated to the past, it is interesting to see how the correlation between ontology and science is making its way beyond theoretical fields, as necessary condition for a shared world based on perception.
It is useful to end with some references to a very peculiar essay written by a philosopher and a geologist, an interesting approach to the science-ontological approach. The essay is written by two professors at the University of Buffalo, Barry Smith and David Mark, and the title is: Do mountains exist? Towards an ontology of Landforms.
This is an odd question. Not so much if you define the context. Indeed, this question leads us to talk about one of the fundamental common sense distinction between appearance and reality. As intuitive and direct as this distinction, there is an immediate answer that leaves no room for uncertainty: “yes sure, mountains do exist”. Mountains exist as a primary correlated of the action and of the human thought. Just think about the importance of spatiality in our actions, our thoughts and mountains play a central role in it.
But mountains also represent the most evident archetype of the geographical object. If we ask ourselves or if we ask a baby or any other person who is learning a new language: tell me a geographical element, most of the times the answer will be “mountain” . Thus, mountains play a twofold role: they can be individually assimilated to objects though in their generality they constitute a natural genre.
Either way you consider them, as individual objects or as a typology, mountains do not exist in the same way as individual objects living in the space we move or our intentions. This is the starting point of the wandering of Smith and Mark who think of matter in many different ways and come to the conclusion that, from both points of view, asking questions equals wanting to find a common language which could do justice to the real presence of mountains, both respecting the scientific aspect of analysis and explanation and the interesting implication emerging from the naive physics studies.