Nicola Spano

AI researchers ask models to describe their own processing, and act on what the models say. Whether such a description is faithful to what the model did has been studied closely. How the question should be put has received far less attention, and no established method says how to put it so that the answer can be analysed rigorously and the analysis repeated. This article proposes such a method: a protocol drawn from the phenomenological tradition, which supplies the interview techniques used in psychiatry and cognitive science. The protocol fixes the interview schedule in advance, asks each question in several wordings, lets no follow-up introduce a word the model has not used, and replaces the binary verdict of faithfulness with a graded one, to be checked against interpretability data. The protocol presupposes nothing about whether these systems are conscious, and it supplies a criterion for deciding whether a model's account of itself is honest. A pilot tested the protocol in ten runs and 896 sessions on two model families, with the predictions of six runs registered before those runs took place. Each session interviewed a fresh instance of the model, meaning one copy started with an empty context. Three of the results bear on any study that questions a model about itself. In the schedule's original order, every instance gave the same answer about its own states; when two of the questions were put in the opposite order, the instances no longer agreed, so the order of the questions can produce what only looks like a finding. A refusal to carry out a task was inserted into an instance's own turn although the model had not written it, and the instance described that refusal as its own. Since every turn carries a label naming who wrote it, a result obtained by prefilling a turn or editing a transcript may follow that label rather than anything the model did. Finally, a pattern in the answers that looked like a report of the instances' own processing appeared just as often when those instances were asked to write in the voice of a fictional character, and two models scoring the same answers under the same written rule gave rates for it that differed by a factor of seven.

In this paper, I argue that according to Edmund Husserl “tendency” does not designate a specific class of intentional experiences but rather, on par with “consciousness-of,” a universal mode of intentionality essential for any constitution of sense. In doing so, I explicate Husserl’s distinction between intentionality as tendency (Tendenz), which he describes as a striving (Streben), and intentionality as consciousness-of (Bewusstsein-von), which he describes as a presentation (Vorstellung) of an intentional object. Then, I discuss Husserl’s problematic way of relating these two universal modes of intentionality. Although he claims that intentionality as tendency presupposes intentionality as consciousness-of, I argue that the universal validity of this presupposition is put into question by the consideration of drives (Triebe), which Husserl describes as passive tendencies that originally lack any consciousness of the end strived toward, and, hence, do not seem to presuppose any presentation of it. I show that the lack of intentionality as consciousness-of poses two major problems in Husserl’s account, in that it makes drives seemingly unintelligible as (i) strivings and as (ii) motivated experiences. Lastly, to find a possible solution to these problems and better clarify the relation between intentionality as tendency and intentionality as consciousness-of at the level of drives, I explore Edith Stein’s account of drives as aimless strivings governed by experiential causality (Erlebniskausalität), discussing its advantages and potential drawbacks, as well as its compatibility with Husserl’s account.

In the present article I aim to make a new contribution to our phenomenological understanding of the foundation between intentional experiences. In order to accomplish this goal, I discuss Husserl’s effort to avoid the conflation of the class of non-objectifying acts, i.e., evaluations and volitions, with the class of objectifying acts, i.e., cognitions. Through the analysis of the transition from his early to his mature account, I explore how Husserl, by readdressing the idea of foundation in relation to the shift from the practical-evaluative to the theoretical attitude, clarifies how evaluations and volitions can exert their intentionality only on the basis of a foundation on cognitions without thereby being reduced to a mere special case of founded representations.

In the present article, I introduce Husserl’s analyses of ‘natural causality’ and ‘volitional causality’, which are collected in the volume ‘Wille und Handlung’ of the Husserliana edition Studien zur Struktur des Bewußtseins. My aim is to show that Husserl’s insight into these phenomena enables us to understand more clearly both the specificity of, and the relation between, the motivational nexus belonging to the sphere of the will in contrast with the causal laws of nature. In light of this understanding, in the last part of the article I reflect on whether and to what extent there is, in fact, an ontological and epistemological compatibility between volitional causality and natural causality.

In this article, I analyze the working principle and associated concepts of the method of eidetic variation as illustrated by Husserl in Experience and Judgment. In so doing, I scrutinize the very specific sense in which, for Husserl, variation must be free in order to have intuition of a universal as pure, that is, as having a priori validity. I then discuss Husserl’s problematic claim that, even if totally free, the eidetic variation is actually not sufficient to achieve purity. The reason for its insufficiency, I argue, must be sought in the habits of phantasy. To this aim, I first undertake a phenomenological analysis of the essence of these habits as passive tendencies of the phantasizing ego. Then, I show that they hinder the intuition of the universal as pure by contaminating the purity of the free possibilities to be varied. Finally, I discuss whether habits of phantasy can be put out of play and whether, depending on the realizability of such a task, the whole method of eidetic variation, and, in relation to it, the possibility of a priori knowledge advocated by Husserl is to be put into question.

In the present article, I address the issue of categorial representative contents, which, according to Husserl’s phenomenological theory, make knowledge possible by providing fullness to intuitive categorial acts. First, I discuss Husserl’s assertion that he no longer approves of his theory of categorial representation developed in Logical Investigations. I argue that the influential interpretation of Husserl’s self-criticism advanced by Dieter Lohmar is unfortunately misleading, as Husserl does not actually claim that categorial representatives are contents of reflection belonging to the realm of inner sensibility, as Lohmar contends. After offering an alternative view of Husserl’s authentic position in Logical Investigations, I examine other possible reasons for his later dissatisfaction. These reasons relate to Husserl’s changing views on the role of categorial representatives in empty thinking acts—necessarily involved in cognition—and the necessity of spontaneously repeating the passive synthesis between the intentional matters that function as categorial representatives in the predicative activity yielding knowledge.

In the present article I discuss, in confrontation with the most recent studies on Husserl’s phenomenology of acting and willing, the taxonomy of action that is collected in the volume ‘_Wille und Handlung_’ of the Husserliana edition _Studien zur Struktur des Bewussteins_. In so doing, I first present Husserl’s universal characterization of action (_Handlung_) as a volitional process (_willentlicher Vorgang_). Then, after clarifying what it means for a process to have a character of volitionality (_Willentlichkeit_), I illustrate the various types of actions, which Husserl distinguishes as ‘straightforward’ (_schlicht_) or ‘deciding’ (_entscheidend_), ‘primary’ (_primär_) or ‘secondary’ (_sekundär_), ‘inner’ (_innere_) or ‘outer’ (_äußere_), ‘immediate’ (_unmittelbar_) or mediate (_mittelbar_), ‘simple’ (_einfach_) or ‘compound’ (_zusammengesetzt_). Finally, I consider Husserl’s discussion of the direction and foundation of action.