The Phenomenology of Semiosis: Approaches to the Gap between the Encyclopaedia and the Porphyrian Tree Spanned by Sedimentation

Abstract

When putting semiotics and phenomenology in juxtaposition, the first task necessarily is to find out what a study of meaning, conceiving of itself as an empirical science, has to do with a philosophical school, the business of which it is to secure the epistemological foundations of all the sciences (broadly understood). Our answer, in short (but we will go at some length to show it), is that since all results of phenomenology also count as contributions to phenomenological psychology, the phenomenological method constitutes a part of the panoply of methods offered to semiotics. Our second task will be to review the fragmentary semiotics proposed, originally employing that term, by Edmund Husserl, to gauge its value for contemporary semiotics. Since our investigation of Husserl’s semiotics will demonstrate that it sometimes concerns the sign in a narrow sense, and sometimes broadens up to a study of meaning in general, our third and final task, in this paper, will be to consider a proposal made by a close follower of Husserl, Alfred Schütz, whose idea of a system of relevancies, wedded to Husserl’s notion of sedimentation, might be amended when considered in connection with Umberto Eco’s idea of the encyclopaedia.

1. Introduction

To begin from the beginning: how it is possible to compare, let alone amalgamate, semiotics, which, at least since the last mid-century, has professed claims to be a scientific study, and phenomenology, the method of which was designed to elucidate some age-old issues afflicting philosophy, and in the process established a sharp-cut opposition to the empirical sciences, and in particular, to psychology. There is indeed one early paper by Edmund Husserl, in which he is explicitly concerned with semiotics, using this very term, but it was written at the time when he still thought psychology could lay the ground for arithmetic [1]. More to the point, in his posthumous papers, Husserl [2] wrote that all results obtained by the use of the phenomenological methods in philosophy should also be valid in phenomenological psychology. Since Husserl’s time, the latter term has been used and abused in many quarters. However, Herbert Spiegelberg [3], otherwise known as the Nestor of the history of phenomenology [4], in a later publication spells out Husserl’s conception in contrast to these later developments in the following terms: it is “an a priori psychology of our inner experience designed to describe its essential structures, analogous to, but still quite different from, pure geometry with its limited system of axioms. /…/ Such a psychology was to provide the basis for empirical psychology.” [3] (p. 11). Some of Husserl’s most direct followers, such as Aron Gurwitsch, Alfred Schütz, and Maurice Merleau-Ponty, were to take this idea further.

Another point needs to be spelled out: to diagnose Husserl’s work as a “metaphysics of presence”, as Jacques Derrida has done, amounts to an “ontologization”, as Burt Hopkins [5] terms it, of something which is really an epistemological approach. It must be admitted, however, that Husserl himself has contributed to a certain confusion concerning what phenomenology is about, by using labels such as “idealism”, “eidetic”, and “transcendental”, which are usually terms involving a brand of ontology. Derrida has admittedly exposed some problems in Husserl’s work, notably, as Steven Crowell [6] observes, pertaining to the nature of signs, but he is unable to bring the phenomenological analysis any further since he declines to argue in phenomenological terms. In fact, Husserl himself struggled all through his life with many of these issues (as we will see in Section 3). It is sufficient to read his posthumous works to become aware of Husserl’s endeavour and to succumb to the temptation to pursue the work.1

2. Phenomenology as a Method in the Empirical Sciences

Phenomenology is a label for a particular method used to describe the invariants of any conceivable experience, so labelled both by Edmund Husserl and by Charles Sanders Peirce (who, however, later changed the name to “phaneroscopy”). Although one can find very similar descriptions of the method in the work of Peirce, Husserl has not only given a more specific description of it, but he has also practised it more assiduously (see [7,8,9]). According to Husserl, phenomenology is a method which aims to describe the content of consciousness as it is experienced, without worrying about whether this content has any basis in fact or not, and then applies to this content the method of variation in the imagination (“ideation”), in order to determine which properties are essential, and which are not, with the ultimate aim of arriving at the characterization of the meaning-type of which the content is an instantiation.

Once we set ourselves the goal of going beyond the very act in which we have our experience in the Lifeworld, Husserl tells us we have to begin by making the epoché, allowing us to put brackets around our immediate experience, not to doubt it, but to suspend judgment concerning its relation to reality, whether it is in the mode of perception, imagination, memory, anticipation, and so on. There follows the phenomenological reduction, by means of which we turn away from the content or object of our experience, which is what counts in ordinary life, being instead directed to (the properties of) the act in which the experience is given.2 There follows the eidetic reduction, which permits us to go from singular experiences to general facts, or, in Husserl’s terms, essences, as a result of ideation, the free variation of features in the imagination. In some of his works, Husserl introduces more steps within the phenomenological method, and, also for other reasons, introduces other terms, but these distinctions will be sufficient for our present purpose.

2.1. The Panoply of Methods: Modes of Access and Phenomena Accessed

Elsewhere, I have situated phenomenology within the panoply of methods which may be used within cognitive semiotics, an approach which aims to pool together the methods, models, and findings of classical semiotics and cognitive science [10,11,12,13,14,15,16,17,18,19,20,21]. In relation to Husserl’s preoccupations, it might be important to note that psychology, together with linguistics, philosophy, and computer science, was an original integrant of what came to be known as cognitive science. Deriving my inspiration from a table designed by Jordan Zlatev [22], which distinguishes first-, second- and third-person methods, I have suggested an amplification of these distinctions, combining different kinds of phenomena studied and different ways of having access to these phenomena, as in

Table 1. The distinction of ipseity, dialogicity, and neutrality in the phenomena accessed and the modes of access (adapted from Sonesson [21] as an extension of Zlatev [22] (p. 1059)). Note that this division into triadic categories is not the same as that made by Peirce, to whom both dialogicity and objectivity would be instances of Secondness. See [23].

		Phenomena accessed
		Ipseity (First person)	Dialogicity (Second person)	Neutrality/Objectivity (Third person)
Modes of access	Ipseity (First person)	Introspection	Empathy	Phenomenology
	Dialogicity (Second person)	Participant observation	Dialogue, Interview	Participant observation
	Neutrality/ Objectivity (Third person)	Behaviouristic description, “hetero-phenomenology”	Interview, Questionnaire	Experimentation Detached observation Brain imagining Computational modelling

[21]. Although I have included similar tables in earlier papers, it was in this latter publication [21] that I decided to employ some less loaded terms, namely, ipseity, dialogicity, and neutrality, to designate the different phenomena as well as the modes of access, in preference to such ambiguous terms as subjectivity, intersubjectivity, and objectivity.

The original reason for amplifying the array of methods was that, in both the cases of introspection and in that of the phenomenological reduction, the mode of access is necessarily the first person, but in introspection, the goal is to ascertain something about the particular person realizing the operation, whereas phenomenology is concerned with mapping out the invariant structures of consciousness, which is to say that it aims to establish some neutral structures. From the point of view of their modes of access, then, both operations rely on ipseity, but only the first operation simply aims to access a phenomenon of ipseity. Indeed, as we shall see later on, the difference between phenomenological “intuition” (not a very felicitous term) and introspection is something on which Husserl, and after him Zahavi and Gallagher, have repeatedly insisted.

There can be third-person reports of the first-person domain of introspection, which is what Daniel Dennett [24], confusingly, calls “heterophenomenology”, whereas a third-person approach which pertains to the third-person domain is called experimentation. Both phenomenology and experimentation aim to attain neutral phenomena, but experimental approaches opt for a neutral mode of access, whereas phenomenology uses an access pertaining to ipseity. In part, experimentation and phenomenology may not be able to attain the same neutral structures. In part, however, they do, but then phenomenology uses an iconic approach, which consists of proceeding to deeper layers of the phenomenon we are exploring, whereas experimentation is reduced to an indexical modus operandi, similar to that in the natural sciences. Since both methods have their intrinsic problems, it is certainly worthwhile employing both approaches (see [25]).

According to Christopher Gutland [26], the phenomenological method simply amounts to introspection, as the latter is defined in The Stanford Encyclopedia of Philosophy [27]. This would seem to contradict the distinction between introspection and phenomenology which I have made above (see Table 1). Gutland [26] (p. 13) goes on to claim, however, that the steps making up the phenomenological method serve to avoid the classical pitfalls of introspection, such as the scope of awareness varying and not being controlled, the influence of prejudices, and the problem of distinguishing between idiosyncratic and general features of consciousness.

The reduction makes sure that the meanings employed in the description are in full concordance with the actual experience. This allows the notice and elimination of false prejudices. The eidetic variation further helps to test claims about necessary structures without being dependent on actual perception. The freedom of this variation helps overcome the limitations of empirical induction. Its results are generalizable, as it makes only indirect use of the possibly idiosyncratic phantasmata. Lastly, intersubjective testing of the results is as important in phenomenology as it is in science (my italics) [26] (p. 13).

Apart from everything else, it is precisely the possibility of distinguishing general facts, or invariants, as Husserl calls them, from idiosyncratic experiences which justifies the distinction between introspection and the phenomenological method (see Table 1). Although the distinction between introspection and the phenomenological method for me grows out of my general familiarity with phenomenology, it turns out that Husserl [28] (p. 36), [29] (p. 38) already insisted a lot on this difference, and so have, more recently, Dan Zahavi [30] and Shaun Gallagher [31].

Still, this does not tell us what part phenomenology can play within an empirical study such as that of semiotics. This is not an issue within Peircean phenomenology, since Peirce never makes a distinction between phenomenology and the empirical sciences. In his late work, as I mention above, Husserl [2] maintained that all the results of phenomenological philosophy will also be valid in phenomenological psychology. Moreover, as Aron Gurwitsch [32] (p. 696) observes, “/e/mpirical psychology is by no means ruled out. Eidetic laws grounded in eidetic structures constitute no more than a framework within which there is room for contingent varieties. Empirical research concerning such varieties, especially the regularities and laws of actual occurrences, is perfectly legitimate under the perspective of eidetic psychology and even called for”.

Presumably, this dictum can be turned around in order to claim that the results of phenomenological psychology are also valid in philosophy, but not to suggest that discoveries within empirical psychology remain in force in phenomenological psychology, let alone in phenomenological philosophy.3 In fact, psychologists, such as Molière’s Monsieur Jourdain, often do phenomenological psychology without knowing it, as I have often observed, particularly in the case of James Gibson (who did know it, however, but did not tell it in his publications: see [33] (pp. 255 ff.)). Moreover, experimental or other empirical studies can suggest regularities, which, with the help of ideation, can potentially be demonstrated to be invariants of human life.

2.2. The Zigzag between Phenomenology and Empirical Studies

A phenomenologist who early on grappled with the conundrum of relating phenomenology to empirical studies was Maurice Merleau-Ponty [34], notably, in his paper Les sciences de l’homme et la phénoménologie (see now Merleau-Ponty [35] (pp. 49–128)). In this paper, after first delineating the radical difference between a phenomenological and an empirical approach to a domain of study, as defined by Husserl, according to which phenomenology must always precede empirical studies, Merleau-Ponty observes that, in actual fact, even Husserl had to admit that psychologists, notably those involved with Gestalt psychology, could not have done what they have done if they had not started out from some kind of phenomenology, however naïve. Merleau-Ponty then goes on to review some passages in Husserl’s late work, which could be taken to suggest that, in the end, Husserl admitted that phenomenology and empirical observation have to proceed in the form of a continuing reciprocal exchange between the former and the latter. Whether or not this is a correct interpretation of Husserl’s late thought or not is not something which we have to decide at present. Although Husserl, from the beginning to the end, speaks of phenomenology as involving a “zig-zag procedure”, it is not clear that this involves any excursion into the “empirical” world (see Orth [36] (pp. 68 ff.)):

Die Untersuchung bewegt sich gleichsam im Zickzack; und dieses Gleichnis paßt um so besser, als man, vermöge der innigen Abhängigkeit der verschiedenen Erkenntnisbegriffe, immer wieder zu den ursprünglichen Analysen zurückkehren und sie an den neuen, sowie die neuen an ihnen bewähren muß (Husserl [37] (p. 17); cf. Husserl [38] (p. 59)). (The investigation zigzags, as it were; and this comparison fits all the better since, due to the intimate dependency of the various concepts of knowledge, one must always return to the original analyses and prove them on the new ones, and vice versa. (My translation, as in the following quotes)).

On the other hand, Husserl’s late work, Krisis, and all the lectures and notes prepared for it, clearly show him initiating his phenomenological inquiry from the point of view of the precise historical situation in which he was situated [38,39]. In any case, this is a conception implemented by Merleau-Ponty, no doubt already in his early work (Merleau-Ponty [40]), but, more meticulously, in his lectures at the Sorbonne in the 1950s of the last century (Merleau-Ponty [41,42]). Or, as David Carr [43] (p. 83; cf. p. 110) observes:

Clearly such understanding is not possible unless the phenomenologist continues in some sense to live in the natural attitude that is being described. Presumably because it is impossible to live in the natural attitude and to observe it phenomenologically at the same time, Husserl often characterizes the pattern of investigation as a zigzag.

Nonetheless, neither phenomenology nor introspection makes much sense, if consciousness is a mere epiphenomenon, a figment of our imagination, or, to use the term employed by Patricia and Paul Churchland as well as by Daniel Dennett, an instance of folk psychology, which will in the near future be eliminated in favour of neurological terms owing to the progress of neuroscience. From this point of view, it does not amount to much of a difference, if you believe, as does Paul Churchland [44], that such folkloristic terms can completely be dispensed with, or if you assume, as does Daniel Dennett [24] (p. 51), that they may still play a part since they are less cumbersome, although it is “bound to be true” that “in principle, neuroscientific levels of description will explain more of the variance, predict more of the “noise” that bedevils higher levels”. In reality, there is absolutely no reason to accept Dennett’s claim. Neuroscientific terms are, in fact, as we saw above (and, more specifically, in [20,45]), only indirect measures of mental experience, and even if we will no doubt continue discovering new ways of obliquely measuring mental life, these indirect measures can never amount to the whole phenomenon. From some points of view, such as discovering causes of mental illness or handicaps, neuroscientific terms can certainly “explain more of the variance, predict more of the “noise””, but that does not mean they are the real thing.

In their fascinating study of the precursor stages to the advent of language in “anatomically modern humans”, Planer and Sterelny [46] reject the distinction “between Gricean and non-Gricean communication” as a basis for studying the difference between human beings and other animals, on the grounds that “/t/he distinction is formulated within the framework of folk psychology” ([46] (p. 231)). As they maintain elsewhere:

If any view like Dennett’s is right, cognitive psychology has been side-lined by framing the evolution of language through the lens of folk psychology. To the extent that folk psychology is a gadget that humans have collectively assembled to interpret one another, it seems unlikely that our own folk psychology is the best framework for describing the mechanisms of minds significantly different from those of contemporary humans ([46] (p. 23)).

Whether or not the kind of folk psychology which disturbs Planer and Sterelny in Grice is the same as that rejected by Churchland and Dennett is a moot question. I have myself had my reasons for rejecting the Gricean notion of communication, for reasons having nothing to do with the status of folk psychology. My problem is that Grice’s distinction simply cannot account for the most common ways in which meaning (“natural” or “non-natural”) is conveyed. Accordingly, I have demonstrated how meaning is ordinarily created and shared (see [47,48,49]). Although they do not refer explicitly to the Gricean model in the following, this is precisely the manner in which Planer and Sterelny go about exposing the ways in which meaning can have been imparted before the advent of sophisticated means of communication such as language. Let us start by admitting that, as folk psychology goes, the Gricean model is already a fairly finespun instance of it, and what I have tried to do, and also what Planer and Sterelny are concerned with doing, is to develop a more refined and precise version of such a folk psychological model. This is precisely what phenomenology is about. In spite of their overt infeudation to the Dennett-Churchland understanding of meaning and mind, Planer and Sterelny never really try to speak neurospeak, not simply because that would be too cumbersome, but because it would fail to make sense, both to themselves and to their readers.

2.3. Naturalization One Way or the Other

So far, we seem to have been talking about what, in recent decades, has been termed “the naturalization of phenomenology”. [50] But, as Dan Zahavi observes in a very lucid paper, there are multiple, and partly contradictory, senses to this term [30,51]. There is a sense in which any naturalization is a “category mistake”, confusing Husserl’s epistemological project with one of the empirical sciences. Even if phenomenology, as the late Husserl recognized, has to take into account the embodiment as well as the social and historical situatedness of the phenomenological subject, the clarification of the very possibility of human knowledge is something distinct from the empirical study of human consciousness. More specifically, it goes against the whole idea of Husserlean phenomenology to posit that our first-person experiences “are ultimately amenable to a natural scientific investigation and explanation” and, more specifically, to a mathematization, which is something directly rejected by Husserl [30] (p. 24). As for the latter point, it might be added that Husserl’s friend Kurt Gödel, whose lasting fame rests with his demonstration of the incompleteness theorem, toyed with the idea that Husserlean phenomenology might constitute a way out of this conundrum [52].

Nevertheless, according to Zahavi’s succinct recapitulation, there is a distinction to be made between what Husserl calls phenomenological philosophy and phenomenological psychology:

For Husserl, the task of phenomenological psychology is to investigate intentional consciousness in a non-reductive manner, that is, in a manner that respects its peculiarity and distinctive features. Phenomenological psychology is consequently a form of descriptive, eidetic, and intentional psychology which takes the first-person perspective seriously, but which—in contrast to transcendental phenomenology, that is, the true philosophical phenomenology—remains within a pre-philosophical attitude and stops short of effectuating the reflective move needed in order to attain the stance of transcendental philosophy [30] (p. 38).

As I observed above (in Section 2.1.), however, Husserl also states that any result in phenomenological philosophy will also be valid in phenomenological psychology. Whether or not this maxim can be turned around is an open question, but it seems fair to admit that, to Husserl, empirical psychology cannot play this part; and yet, it is difficult to avoid the impression that this is what occurs with Husserl’s, let alone Gurwitsch’s and Merleau-Ponty’s, recourse to Gestalt psychology and other empirical approaches to consciousness. Nonetheless, as I also observed, at least in the case of Husserl, it may be more correct to consider these passages to be references to the unacknowledged eidetic considerations of these psychologists, and thus, to phenomenological psychology. But then, it would seem to mean that the results of phenomenological psychology can be recuperated by phenomenological philosophy.

Rejecting the sense of naturalized phenomenology as understood by Petitot et al. [50] that goes directly against Husserl’s professed goal in devising the phenomenological method, Zahavi points to two legitimate ways of “naturalizing” phenomenology. The first of these alternatives is, again, very clearly delineated by Zahavi:

Empirical science can present phenomenology with concrete findings that it cannot simply ignore, but must be able to accommodate; evidence that might force it to refine or revise its own analyses. At the same time, phenomenology might not only contribute with its own careful descriptions of the explanandum, but might also question and elucidate basic theoretical assumptions made by empirical science, just as it might aid in the development of new experimental paradigms [30] (p. 35).

Unlike Zahavi, I have no doubt in my mind that this approach should be considered a “desideratum”, not for the sake of phenomenology, but for that of the semiotic (human and social) sciences. It is, in fact, the way I have used phenomenology from my first publications onwards (Sonesson [33,53]). Indeed, this approach also implies preparing for the experimental layout, or whatever empirical method is to be used, by phenomenological analysis, in order to ascertain what variables may be relevant, and to evaluate the results of the study using phenomenology. Although my own notion of this “dialectic of phenomenology and experiment” was formulated quite independently, it seems to correspond roughly to what Gallagher calls “front-loaded phenomenology” [54] (p. 91), with a misleading formulation which I have repeatedly criticized, but for which the same author has also found a more felicitous formulation: “phenomenologically informed experimental design” [31] (p. 125).

The second legitimate approach which Zahavi [30] (p. 41) recognizes involves “a reexamination of the usual concept of naturalization and a revision of the classical dichotomy between the empirical and the transcendental”. The same approach, it would seem, is what Gallagher, curiously, terms “experimental phenomenology” [54] (p. 130). Such a task, as Zahavi rightly observed, is staggering. And yet, this is certainly the task undertaken by Merleau-Ponty, and, as Zahavi seems to acknowledge, by the late Husserl. In the present paper, nevertheless, I will try not to stray from the first of Zahavi’s legitimate approaches.

One is easily misled when reading Zahavi’s papers on naturalizing phenomenology to believe that naturalized phenomenology is the same as “neurophenomenology”. Even if we grant the distinction made by Zahavi between neurophenomenology in the narrow sense, as envisioned by Francisco Varela and realized by Antoine Lutz [55], and a wider sense of the term, its import cannot cover all varieties of naturalized phenomenology. In the first sense, the term is disingenuous: in the case of the experiment in which people trained to use the phenomenological method are submitted to brain imagining during the process, a more apt term for what is going on would be a neuroscientific study of the very quaint mental state elsewhere termed the phenomenological method. This is no doubt an incomplete characterization of what Varela meant by the term neurophenomenology and even of what Lutz [56] and others use the term to mean nowadays, but it does cover a central procedure involved. As for using the term more widely to encompass “various attempts to naturalize phenomenology, attempts that are all guided by the idea that (philosophical) phenomenology and empirical science are mutually constraining and enlightening projects” [30] (p. 27), it is simply a misnomer. As we have seen above, there are many other methods by means of which phenomenology may encounter empirical studies.

A deeper terminological issue, nonetheless, is whether we should still talk about a “naturalization” of phenomenology, once we are not concerned with reducing the phenomenological results (as is the intention of the authors of the naturalization volume [50]) to the terms of the natural sciences. Since our compromise is, on the contrary, to consider meaning to be the “real thing”, of which quantitative measures can only offer indirect adumbrations, we might have a better talk about semiotization. In the following, nonetheless, we will talk about phenomenological psychology or, for short, phenomenology.

3. The Semiotics of the Sign and Other Meanings

In many of my earlier publications [19,33,57,58,59], I have proposed a minimal definition of the sign, in which I have attributed the first part of my criteria to Husserl (the double asymmetry), and the second part to Jean Piaget (subjective differentiation). The definition is minimal, because more criteria may have to be added in order to circumscribe the prototypical notion of the sign (cf. Sonesson [58]).

The sign contains (at least) two parts (expression and content) and is, as a whole, relatively independent of that for which it stands (the referent):

• There is a double asymmetry between the two parts, because one part, expression, is more directly experienced than the other;
• And because the other part, content, is more in focus than the other;
• These parts are differentiated from the point of view of the subjects involved in the semiotic process, even though they may not be so objectively, i.e., in the common sense, Lifeworld (except as signs forming part of that Lifeworld);
• This means that the two objects serving as expression and content do not go over into each over without rupture, contrary to what happens in perception;
• This also means that the two objects are experienced as pertaining to different categories of Lifeworld experience;
• The sign itself is subjectively differentiated from the referent, and the referent is more indirectly known than any part of the sign (as a general experience of the Lifeworld, not necessarily at the moment of experiencing the sign).

In more recent times, I have begun to have my doubts about whether this constitutes a well-founded hermeneutic of Husserl’s position, although I have other reasons for thinking that it amounts to a fairly adequate phenomenology of the sign (see [13,50,51]). In this section, I intend to go back to the sources, in order to explore whatever foundation there can be for my claim to be following Husserl in announcing the first part of the criteria for the sign definition, as well as to what extent the second part of the criteria can also be prefigured in Husserl’s work.

3.1. The Semiotics of Edmund Husserl

If we except the draft (not the published version) of his review of the Logik of Ernst Schröder [60] (pp. 381–299), Edmund Husserl [61] seems to have used the term “semiotics” only once, as the subtitle to a text written around 1890, the main title of which is “Logik der Zeichen”. This text has been curiously neglected in the secondary literature until recently (but now, see [62,63,64,65,66,67,68,69,70]). I did refer to it in my dissertation [52], but admittedly, in a rather superficial manner. In both these texts, Husserl uses the term “sign” in a very extensive way: “Das Wort Zeichen in unserer Definition ist in dem denkbar weitesten Sinn zu nehmen.“ (“The word sign in our definition should be taken in the widest sense imaginable”. My translation, as in the following quotes). Not only should the term apply to nouns generally, as well as names, but to any feature of an object: “Des weiteren gilt uns /…/ jedes begriffliche Merkmal, sofern es eben als Merkmal dient, als Zeichen. Jedwede Beschaffenheit, sei es eine absolute oder relative, kann gelegentlich als Merkzeichen dienen des Gegenstandes, der sie besitzt.“ [61] (p. 340f.) (“To us, furthermore, /.../ every conceptual feature, insofar as it serves as a feature, counts as a sign. Any quality, whether absolute or relative, can occasionally serve as a marker of the object that possesses it.”). The advantage of using the term “sign” in such a broad fashion is that it amounts to conceiving semiotics as the study of meaning generally. The disservice done by such a wide notion of “sign”, however, consists of treating all meaning on a par and not singling out the specific meaning of the sign, which is the notorious problem I have pointed out elsewhere with the Peircean notion of sign, as well as with the practice of the French structuralists (see [8,19,33,57,58,59]). Perhaps “cue” would be a more adequate translation for what Husserl here is concerned with. But, it could also be seen as an anticipation of the latter term “appresentation”, as one of the many meanings it takes on in Husserl’s work (see below).

In retrospect, we recognize in the Schröder review a formulation which seems to anticipate a persistent theme of Husserl’s phenomenology while using the term semiotics: “Wie es möglich ist, daß ein blinder Mechanismus von sinnlichen Zeichen logisches Denken ersetzen und ersparen kann, das ist die große Frage der Logik der Zeichen, der Semiotik“ [60] (p. 394). (“How it is possible for the blind mechanism characteristic of sensory signs can replace and economize logical thinking, that is the great question of the logic of signs, of semiotics”). Indeed, in the Semiotik text, where several sets of sign divisions are proposed, there is one which stands out: that between “uneigentliche Vorstellungen”, which serves as proxy for “eigentliche Vorstellungen” and which eventually may be reached, and those which fail to be attained, either because it is more convenient to hold on to the former, or because there is no way of grasping the latter.

Die uneigentlichen Vorstellungen können nämlich: (1) als bloße Vermittler zur Erzeugung der ihnen korrespondierenden eigentlichen Vorstellungen dienen. In dieser Art funktionieren z. B. konventionelle Abzeichen, mnemotechnische Wortfolgen, mechanisch eingelernte Verse und dgl. (2) Die uneigentlichen Vorstellungen können aber auch als Surrogatvorstellungen die eigentlichen ersetzen [61] (p. 351). (The inappropriate ideas can, namely: (1) Serve as mere intermediaries for the production of the real ideas corresponding to them. Conventional emblems, mnemonic word sequences, mechanically learned verses, and the like, for example, function in this way. (2) The inauthentic ideas can, however, also as surrogate ideas, serve as replacements for the real ones).

The case in which the “real notions” are neglected because it is commodious to use some abbreviation, and the one in which the “real notions” are unavailable, as exemplified by the series of natural numbers, which is infinite, but also, for instance, by the notion of God, are both called surrogates (see [67] (pp. 190ff.), [62,69] (pp. 218ff.)). Although Husserl later was to abandon the surrogate theory, it is easy to recognize here what later became the difference between the intuitive idea and its fulfilment, notably in perception.

In Logische Untersuchungen, on the other hand, Husserl [37] resorts to a much narrower sense of the term “sign”, or rather to two different notions of sign. Following the lead of Bernard Bolzano, he even claims that these two notions of sign have nothing in common (see Majolino [66]). Thus, he seems to be undoing the synthesis operated by Augustin between the Stoic and the Aristotelean senses of sign, which, according to John Deely [71] (p. 17f.), was defined as “a thing which, over and above the impression it makes on the senses, causes something else to come into thought as a consequence” (see also [58,72,73,74,75]). The Stoic sense of the sign, exemplified, as Giovanni Manetti [76] has shown, by such age-old practices as divination and medical symptoms, and no doubt also by that of game hunting, conforms to the formula that if something is the case (p), then something else is also the case (q). We can call this sense of sign indication or inference. To Aristotle, on the other hand, a sign is basically a linguistic sign, in which the written word stands arbitrarily for the spoken word, which in turn stands arbitrarily for the mental content, which itself, since it is said to be the same for all human beings, can be taken to stand for the same object in the world for all of them.4 This notion of sign is obviously more akin to the notion of sign which we know from Saussure, and from the philosophers of the Ideological school before him. In Mutatis mutandis, this is the same distinction made by Bolzano, and it reappears, in a somewhat different form again, in Logische Untersuchungen.

To Bolzano, these two notions of sign are different, notably because indication, contrary to the other sense of sign, proves that the object in question exists. I have argued elsewhere that this is an erroneous interpretation (see [77]): indication is really an act of interpretation, but while the latter can be ascribed to a particular agent, the former is the product of the knowledge shared by a community, which is operated on the spot by the receiving agent, making the community and the addressee co-addressers of the act. Thus, such as in the case of divination, and certainly also that of medical symptoms, it may easily point to an equivocal or non-existing object. Husserl’s use of the notion does not seem to be vulnerable to this rebuke, since he (possible influenced by Franz Brentano: see Majolino [66]) merely supposes that the conviction that a certain state of affairs obtained motivates the conviction (which may even be unreasonable) that another state of affairs equally obtains:

daß irgendwelche Gegenstände oder Sachverhalte, von deren Bestand jemand aktuelle Kenntnis hat, ihm den Bestand gewisser anderer Gegenstände oder Sachverhalte in dem Sinne anzeigen, daß die Überzeugung von dem Sein der einen von ihm als Motiv (und zwar als ein nichteinsichtiges Motiv) erlebt wird für die Überzeugung oder Vermutung vom Sein der anderen (Husserl [37] (p. 25)). (that any objects or states of affairs, of the existence of which someone has current knowledge, indicate to him the existence of certain other objects or states of affairs in the sense that the conviction of the existence of one is experienced by him as a motive (and indeed as a not explicitly understood motive) for the belief or assumption of the existence of others.).

More importantly, he broadens the notion of indication (Anzeichen) to include all kinds of inferences, which means that his other category of sign, expression (Ausdruck), must, at the same time, also be an instance of indication. He concludes that:

alle Ausdrücke in der kommunikativen Rede als Anzeichen fungieren. Sie dienen dem Hörenden als Zeichen für die „Gedanken” des Redenden, d. h. für die sinngebenden psychischen Erlebnisse desselben, sowie für die sonstigen psychischen Erlebnisse, welche zur mitteilenden Intention gehören (Husserl [37] (p. 33)). (all expressions in communicative speech function as indications. They serve the listener as a sign for the “thoughts” of the speaker, i.e., for the meaning-giving psychic experiences of the speaker, as well as for the other psychic experiences which belong to the communicating intention. (My translation)).

Indeed, Husserl mentions as cases of indication not only features of an object which allow it to be identified, but also the brand on the slave’s body, the flag as standing for the nation, the channel on Mars as a sign of intelligent life, the fossils as signs of earlier living creatures, and the knot on the handkerchief made to remember something, as well as monuments (Husserl [37] (p. 24)). It stands to reason that this list contains signs in both Bolzano’s sense of the term, in other words, both Stoic and Aristotelean signs. Yet, Husserl also claims that the term sign in ordinary parlance stands for two very different things.

The specificity of the expression (Ausdruck) in Husserl’s sense resides elsewhere (see [66]). To begin with, it must be verbally formulated, or be similar to language in some not specified sense. It involves:

jede Rede und jeder Redeteil, sowie jedes wesentlich gleichartige Zeichen ein Ausdruck sei, wobei es darauf nicht ankommen soll, ob die Rede wirklich geredet, also in kommunikativer Absicht an irgendwelche Personen gerichtet ist oder nicht (Husserl [37] (p. 30f.)). (every kind of speech and every part of speech, as well as every essentially similar sign, being an expression, whereby it should not matter whether the speech is actually spoken, i.e., addressed to any person with communicative intention or not.).

Whatever “gleicharte Zeichen” means, it does not include gesture and facial expressions, because they are not part of the “mitteilender Absicht”. So far, it may seem, in retrospect, that Husserl is going to arrive as some kind of “non-natural meaning” in the sense of Grice (see [37] (pp. 39f.); [78] § 32). Husserl’s next step, however, is to concentrate on the case in which language is used without any communicative intent, that is, as a kind of inner monologue, no doubt because, as he observed, any expression used in communication is necessarily also an indication, minimally, with reference to the consciousness of the one engaging in communication. Husserl [37] (pp. 23ff.) is certainly more interested in the notion of expression than that of indication, no doubt because he (unlike Peirce) thinks that logic, i.e., propositions, can only be formulated by means of language, and that language requires signs which are explicitly and purposefully manifested by a responsible subject—which is, roughly, the meaning of expression to Husserl at this time.

This kind of soliloquy (not to be identified with the “inner dialogue” of Vygotsky and Lotman) is what has spurned the famous critique by Jacques Derrida [79] of Husserl’s “metaphysics of presence”. In Derrida’s interpretation, this is a case of the ego’s presence to itself. To make sense of this interpretation, we must place it in the context of one of Husserl’s most recurring themes, the dialectic (using a non-Husserlean term) of presence and absence. What Husserl declared, in the Schröder review [60] (p. 394), to be the fundamental issue of semiotics, as the logic of the sign, runs through his whole work as the opposition between intuition and its fulfilment (Erfüllung). This is a kind of dialectic which applies primarily to perception, and thus, only secondarily to the kind of indirect perception which can be conveyed by signs.

3.2. Beyond the Surrogate Theory of the Sign

In his early semiotics text, Husserl clearly defends a conception of the sign as a surrogate, to the point of positing a kind of similarity between signifier and signified even in language. Such a conception would seem to accord with the mediaeval formula aliquid stat pro aliquo, more recently made popular by Roman Jakobson [80]. Already in Logische Untersuchungen, nevertheless, he unequivocally rejects this conception. As Thomas Byrne [62] (p. 225) observes, “one gravely misguided philosopher” (that is, Derrida [79] (p. 57)) diagnosed an obstinacy in Husserl to preserve presence, when, in reality, Husserl’s accomplishment, in abandoning the surrogate theory of signs which he had inherited from Bolzano and his predecessors, consisted of the recognition that it is basically the intuition of an object, outside of its presence, which gives meaning to the object, which then is fulfilled by the presence of the object or not. However, it may be more accurate to observe, as does Robert Sokolowski [81,82], that “presence and absence”, together with “parts and wholes” and “identity in a manifold”, form the three deep-rooted formal structures of phenomenology.

In his remarks for the revision of the case studies in the second part of Logische Untersuchungen, Husserl (as quoted in Sinigaglia [83]; but now see [84]) claims that only Ausdrücke are really signs (Zeichen), because indications (Anzeichen) only mean anything to the extent that they carry meaning:

Die Bedeutungsfunktion ist nämlich nicht mehr nur Vorrecht des sprachlichen Ausdrucks: Husserl meint nun, daß man auch im Fall der Erinnerungszeichen, der Merkzeichen und der Signale usw. von echten Zeichen sprechen kann. Mit ihnen ist etwas gemeint, und zwar: “mit dem Stigma ist gemeint: Das ist ein Sklave. Mit der Fahne ist gemeint: Das ist ein deutsches Schiff. Mit dem Sturmsignal: Sturm ist im Anzug” (Ms. A I 17/II, BL. 57b). [83] (p. 192ff.) (The meaning function is to wit no longer only the prerogative of linguistic expression: Husserl now means that one can also speak of real signs in the case of reminders, markers, signals, etc. Something is meant by them, namely: “by the stigma is meant: This is a slave. The flag means: This is a German ship. With the storm signal: Storm is approaching”).

It would seem, then, that if Husserl had gone through with the changes that he had contemplated for the new version of Logische Untersuchungen, he would have ended up with a new synthesis of the kind proposed by Augustin. Taken together with Husserl’s rejection of the surrogate theory of the sign, this implies that he would have rejected the broader notion of sign elaborated by such thinkers as Pedro da Fonseca, who was active in Coimbra on the Iberian Peninsula in the 16th century, according to whom a sign is anything which serves to bring into awareness something different from itself, whether the sign (in the sense of the signifier) itself becomes subject to awareness in the process or not (see Deely [71] (pp. 52ff.); [74] (pp. 414ff.); [85] (pp. 58ff.)) Or, seen from another point of view, which would clearly be that of Husserl, we consider this a case in which there is no sign, but an object carrying its own meaning, that is, an intentional object.

3.3. Signs and Appresentations

Alfred Schütz [86] (p. 148), who in many other respects must be considered a close follower of Husserl, begs to differ from “the rather fragmentary manner” in which Husserl discussed the theory of sign in Logische Untersuchungen, maintaining instead that: “Husserl’s theory of appresentation as developed in his later works can be fruitfully applied to the relationship between the sign and the significatum”. In a later paper, he goes on to propose a quadruplicate layering of pairings making up the sign in the following manner:

In general, we may state that in any appresentational situation the following four orders are involved: (a) the order of objects to which the immediately apperceived object belongs if experienced as a self, disregarding any appresentational references. We shall call this order the “apperceptual scheme.” (b) the order of objects to which the immediately apperceived object belongs if taken not as a self but as a member of an appresentational pair, thus referring to something other than itself. We shall call this order the “appresentational scheme.” (c) the order of objects to which the appresented member of the pair belongs which is apperceived in a merely analogical manner. We shall call this order the “referential scheme.” d) the order to which the particular appresentational reference itself belongs, that is, the particular type of pairing or context by which the appresenting member is connected with the appresented one, or, more generally, the relationship which prevails between the appresentational and the referential scheme. We shall call this order the “contextual or interpretational scheme.” [86] (p. 298).

The idea of appresentation as, among other things, a building block of signs, was developed by Alfred Schütz [87] (pp. 207ff.) and specified by Thomas Luckmann [88]. In Husserl’s ([89] (pp. 174ff.); [90] (pp. 238ff.)) parlance, different phenomena form a paired association, or a coupling, when both items are directly present; they constitute an appresented pairing, or simply an appresentation, when one of the items is present and the other is not; and an appresentation becomes a sign when it is the absent item which is the theme (cf. Luckmann [88] (pp. 205ff.)). But let us have a closer look at what Husserl means by the term appresentation (Appräsentation). This is a term which appears very rarely, as far as I can tell, in Husserl’s works. One such instance explains it as: “der Appräsentation, der Mithabe als mitseiend mit einem original Gegebenen” [91] (p. 422). Otherwise, the term mainly seems to play a part in Cartesianische Meditationen [92], where it is applied to our experience of other subjects, which is characterized as: “eine Art des Mit-gegenwärtig-machens, eine Art Appräsentation”. The problem which Husserl discusses in the following, however, is how the experience of the other can be an appresentation:

Eine solche liegt schon in der äußeren Erfahrung vor, sofern die eigentlich gesehene Vorderseite eines Dinges stets und notwendig eine dingliche Rückseite appräsentiert, und ihr einen mehr oder minder bestimmten Gehalt vorzeichnet. Andrerseits kann es gerade dieße Art der schon die primordinale Natur mitkonstituierenden Appräsentation nicht sein, da zu ihr die Möglichkeit der Bewährung durch entsprechende erfüllende Präsentation gehört (die Rückseite wird zur Vorderseite), während das für diejenige Appräsentation, die in eine andere Originalsphäre hinein leiten soll, apriori ausgeschlossen sein muß [92] (p. 139). (Something like that is already present in external experience, insofar as the front side of a thing that is actually seen always and necessarily apprepresents a real back side and prescribes a more or less specific content for it. On the other hand, it cannot be question of precisely this type of appresentation, that is already co-constituted in primordial nature, since it includes the possibility of validation by means of a corresponding fulfilling presentation (the reverse side becomes the front side), while this must be excluded a priori for the appresentation that is intended to lead into another original sphere.).

Here, we are not interested in how Husserl resolves this paradox, but what he here presupposes to be the original nature of appresentation: it is something which accompanies a presentation, such as a percept, and which itself can be transformed into a presentation, for example, when the object of a perception is turned around, so that other sides of the object become visible. To Husserl, appresentation involves the co-experiencing of the hidden sides of the object. “Every perception simultaneously presents and appresents. It appresents the empty horizons around the direct perception” [93] (p. 40). (See also [94] (p. 40)). Appresentation can also apply to the temporal horizons, that is, retention and protention [94] (p. 40). In this sense, appresentation constitutes an immediate experience. Or, at least, it is almost immediate: it is only a question of turning the object around or waiting for time to elapse. It is not a notion which serves to define the sign.5

It can now be seen that my definition of the sign, in respect to which it depends on Husserl’s phenomenological investigations, involves the Anzeichen rather than the Ausdruck, and that, more generally, it is inspired by the dialectics of intuition and fulfilment (Erfüllung), which characterize all experience in Husserl’s view. Indeed, it comes closest to the notion of Zeichen, as in Husserl’s projected revision of Logische Untersuchungen mentioned above. Even though the formulations in this definition may give the impression that there is one object which takes the place of another, I have already elsewhere rejected the surrogate theory of the sign, without having taken cognizance of Husserl’s veto. Still, these formulations are inevitable, for signs, also known as representations, should not be confused with presentations (including appresentations), let alone with presentifications.

3.4. Presentations, Representations, and Presentifications

In his labyrinthine investigations into fantasy, memory, and pictures, Husserl [95] tries out many distinctions, but there is one which I have found particularly enlightening, since it clearly distinguishes “eigentliche Vorstellungen”, which, apart from perception includes memory, anticipation, and “mental images” (to the extent that they are not really images), and “uneigentliche Vorstellungen”, which divides into pictorial experiences and signitive experiences [95] (p. 139). Elsewhere, I have suggested that a more explanatory terminology would oppose experiences based on single intentionality and those which require multiple intentionalities [19]. Both presentations and presentifications are single intentionality experiences. Representations, whether they are signs or pictures, require (at least) double intentional objects. This is the sense in which “object” should be understood in the definition above.

As Aron Gurwitsch [96] (pp. 176f.), has observed, in phenomenological parlance, perception is said to carry meaning “in a more broad sense than is usually understood”, the latter being often “confined to meanings of symbols”, that is, our signs. In this wider sense, meanings may form parts of other meanings, as (potential) appresentations of the whole that is immediately perceived, which is different, in that respect, from signs in the case of which the signifier is not part of the signified, or vice versa. This could be taken to suggest that signs are made up of (at least) two objects—indeed, two intentional objects (as suggested in Section 3). Indeed, as Gurwitsch [96] (pp. 262ff.) goes on to pinpoint, meaning is already involved in the perception of something directly perceived as marks, or sounds, which then serve as carriers of meanings found in words. He faults psychologists with being subject to this confusion, apparently being unaware of James Gibson [97,98] having made the same distinction between direct perception and referential meaning.

Thus, as indicated by Husserl’s [95] (p. 139) distinctions reproduced in Figure 1, and as I have developed them elsewhere [99], simple or direct intentionality is sufficient to account for all presentations (including appresentations in time and in space) and all presentifications. The latter acts merely to make present something otherwise present, but in an altered way. This is different from the true representations, which are based on multiple or mediated intentionality. Here, we encounter again the dialectic of presence and absence, which Robert Sokolowski [81,82] has rightly described as one of the three basic dichotomies of phenomenology.

However, although they are presentations, there is a sense in which retentions and protentions, as well as the hidden or adumbrated sides of things observed in the here-and-now, are, if not absent, at least less present than the object focused within the precept. Retentions and protentions are the moments in the stream of consciousness which immediately precede and follow, respectively, the moment of our present attention. Retentions include earlier retentions, and so on interminably. Protentions, in the same way, include later protentions without fixed limits. And there may be retentions of protentions, and protentions of retentions. Since the stream of consciousness does not stop, all protentions and retentions must, at some moment, lose their identity as such and transform into the now point. The issue of the adumbrated and hidden sides may be more troublesome. If I hold an object in my hand and turn it around, so as to inspect its other sides, I am clearly initiating an act, by means of which the adumbrated and hidden sides eventually transform into the here point. The hidden and adumbrated sides still remain parts of the totality of the object experienced. But, in order to do so, I have clearly put into action a process which engenders new retentions and protentions.

4. From Relevancies to the Rhizome

If retentions and protentions, as well as hidden or vaguely adumbrated sides of objects, are presentations, then it is not always clear, in our real-world, situated experiences, when something is a sign or just a kind of appresentation which is not currently directly facing us in experience. Or, to have recourse to another Husserlean terminology, the inner and outer horizons clearly contain meanings which are not as directly given as that which confronts us here and now. Dan Sperber and Deidre Wilson [100] understandably have rejected what they, misleadingly, call “the semiotic model”, in favour of a model based on what they term “relevance”. Nonetheless, they never explain what they mean by this term, which is why I have elsewhere [49] taken an interest in the more explicit model of “systems of relevancies” [101] first delineated by Alfred Schütz [102].

In his reflection of relevance, Schütz [102] (pp. 4ff.) tells a story about Carneades, who, when entering a room which is badly lit, wonders whether what he sees in the corner is a pile of rope or a coiled snake. Initially, he is roughly equally motivated to believe the object to be one or the other (see [102] (pp. 16ff.)). Carneades then realizes that the object is not moving, which offers him some elementary evidence for taking it to be merely a coil of rope. Continuing the inspection of the object, however, he is reminded that it is currently winter, and that snakes are torpid at this time of year. Finally, he picks up a stick, strikes the object in question, and observes that it still does not move, thereby corroborating the interpretation of it as a coil of rope. He has not contented himself with gaining evidence at one level, but has sought out additional indications and counter-indications which could pertain to the situation. Thus, Schütz turns a sceptic’s argument into a narrative of our progressive search for truth, which we can approach ever further without definitively attaining it, as both Husserl and Peirce have observed (see Sonesson [49]).

4.1. Carneades at the Inn: The Case for Sedimentation

Schütz [102] explicates what happens in this story by recurring to the notion of a “system of relevancies”, all of them, broadly speaking, social in nature, and having the function of guiding our interest in given situations as they occur in the Lifeworld (see Sonesson [19]). While he 102] (pp. 25ff., 30ff.) does not forget about the contingencies of the present situation, the main thrust of his argument consists of imputing relevancies to the typicalities of the Lifeworld, in Husserl’s sense of the term. In Schütz’s terminology, the Carneadean man has sought out those perceptions and sedimented experiences from his stock of knowledge which are relevant to the problem at hand. As Schütz goes on to observe, Husserl [89] might well be able to go along with this description, as far as it goes, calling it a case of problematic possibilities, but he would point out that the situation into which the Carneadean man treads is a kind of prepredicative experience, preconstituted by passive synthesis, which accounts for the experience of similarity, likeness, contrariness, and all kinds of typicalities of which the situation is largely made up. Such elements, I submit, are also what make Schütz’s analysis of this situation different from those proposed by Paul Grice [103] and his latter-day apostates Dan Sperber and Deidre Wilson [100] (see [41].) The case will be different, Schütz points out, whether the Carneadean man enters his own bedroom, where everything, except for this object, is stamped with the property of familiarity, or whether instead, he enters a room where he has never been before, where in principle everything could be new to him. Now, suppose what he enters is a bedroom, either at a hotel, or in the apartment of some friends who have convinced him to stay overnight for the first time. In this case, basically everything will be new to him, but on the level of instances, not on the level of types. He (if we transport him to more recent times) will expect there to be (at least) one bed, complete with bedclothes, and very probably one or two bedside tables, one chair or more, a few lamps, at least in the ceiling, and (at least in the case of the hotel) a desk. In the latter case, he would nowadays expect there also to be a telephone, a television set, and a Wi-Fi connection.

In his main work in German, before he had to emigrate, Schütz [104] abundantly referred to the notion of schemes of interpretation (Schemata unserer Erfahrung, and also, Deutungsschema), which he presented as the result of earlier sedimented acts of experience, which formed the framework on the basis of which present acts acquired their meaning:

Ein Schema unserer Erfahrung ist ein Sinnzusammenhang unserer erfahrenden Erlebnisse, welcher zwar die in den erfahrenden Erlebnissen fertig konstituierten Erfahrungsgegenständlichkeiten erfaßt, nicht aber daß Wie des Konstitutionsvorganges, in welchem sich die erfahrenden Erlebnisse zu Erfahrungsgegenständlichkeiten konstituierten. Das Wie des Konstitutionsvorganges und dieser selbst bleibt vielmehr unbeachtet, das Konstituierte ist fraglos gegeben [104] (pp. 87f.). (A schema of our experience is a meaningful context in the experiences we live through, which indeed comprehends the experiential objects as being fully constituted in the experiences encountered, but not the How of the constitutional process in which the experiences constituted gone through are constituted as experiential objects. Rather, the How of the process of constitution and the process as such remain unnoticed; what is constituted is unquestionably given.).

In his later work in the United States, Schütz never explained in what way what he then called systems of relevancies relate to his earlier notion of scheme. Perhaps these systems of relevancies might be conceived to be made up of schemes, or as being equivalent to schemes, in which case we have a least something more of an account of the passive synthesis behind it, in other words, of the processes of sedimentation. The notion of sedimentation invoked by Schütz [104] (pp. 33, 81, 84, 107) stems from the late work of Husserl, according to whom [38] any present act of experiencing an object or state of affairs is embedded in patterns of understanding which modify these experiences, resulting from the process of sedimentation. This is the process in which previous experiences come to shape and condition more recent ones (see Figure 2). Husserl’s conviction is that such an accumulated product of experience can be reanimated in the phenomenological process, thus illuminating its validity in the sense of its foundation. In posthumous texts, Husserl distinguished between the genetic and generative dimensions of experience (Husserl [105,106,107], also see [108,109]); in the first case, the layering, or sedimentation, is connected to acts having their origin in our personal experience, which is what gives them their validity, whereas the second case results from the layering, or sedimentation, of the different acts in which they have become known from the experience of earlier generations of human beings, deriving from historical processes of collective experience over time.

It is important to note that the approach in terms of geneticity and generativity supposes accumulation/sedimentation to be as much a result of communication and vice versa. This does not only apply to semiotic acts, but to all acts accomplished by situated subjects. In other terms, each act of communication (and of meaning generally) adds to the sedimentation resulting in the pool of knowledge, and each act is also a realization of such a pool of knowledge (see Figure 2). If we admit that each act of meaning must somehow be realized in experience, we can apply the term enunciation in a broader sense to also include, relatively speaking, more passive experiences such as perception. The situation of enunciation is, thus, where types are turned into tokens, and tokens into types. This itself is a result of the sedimentation, the passive (and sometimes active) synthesis of earlier acts (tokens) which, throughout history, have formed Schutzian schemes of interpretation/systems of relevancies when applied to specific situations.

4.2. A Protracted Stay at the Inn: Signs and Meanings

Also, Aron Gurwitsch uses the term “relevance”, notably, in the version of his most famous book first published in French, where a more correct translation of the English term should have been “pertinence” (see Gurwitsch [94] (p. 271)). In French, the term “relevance” does not exist, though it may later, at the time of French structuralism, have infiltrated French scholarly discourse coming this time from linguistics. Nevertheless, there is a French verb, “relever”, which, among other things, signifies something like “depending on” or “pertaining to a particular domain” (Le Petit Robert: “être du ressort de, dépendre de, être du domain de”). Indeed, this is precisely the meaning given to the term by Gurwitsch ([110] (p. 270), [96] (p. 340); his italics), that which is relevant is not simply co-present with the theme, but it is: “of a certain concern to the theme. They have something to do with it.” Nevertheless, Gurwitsch ([96] (p. 342)) observes that, “though occasionally using the term in a sense close to ours”, Schütz seems to understand relevance much more with reference to a given, embodied, and situated Ego. This seems to me less true about Schütz’s later writings, taking into account his recourse to the Husserlean notion of typicality.

In spite of Gurwitsch’s critique, I think we are justified in seeing in Schütz’s relevancies a kind of thematic adumbration. At least Schütz’s [102] (p. 26) topical relevancies could be understood in this sense: as “that by virtue of which something is constituted as problematic in the midst of the unstructuralized field of unproblematic familiarity—and therewith the field into theme and horizon.” From a Gurwitschean point of view, nevertheless, one may wonder for whom something becomes problematic while other things remain familiar. The interpretational relevancies seem to involve the different possible interpretations of what the problematic item could turn out to be, which, in the Carneades case, may be a pile of rope or a snake, and perhaps other things, but certainly not a table or a bed [96] (pp. 38ff.). These interpretations seem to me to be difficult to separate from the topical relevancies, of which they rather form a part, somewhat like a paradigm, a set of alternatives, as put in relation to a syntagm, the chain of connected items. The motivational relevancies are more obviously beside the point in a Gurwitschean perspective, because they have to do with the motives which make us act on our interpretations [96] (pp. 45ff.). But, Schütz might have been better inspired to treat topics, interpretations, and motives as different aspects of relevance systems. It is in this sense that relevancies can contain that which is given as well as that which is new. As I have observed elsewhere [49], Jean-Louis Dessalles [111], who takes his cues from Sperber and Wilson, nevertheless, understands relevance as the opposite of that which is taken for granted, that is, that which introduces some new element to the situation. A system of relevancies, however, may comprehend both aspects: that given is that which is accounted for by the scheme of interpretation. The new part is that which stands out from this given interpretation.

Let us now suppose that Carneades has stayed at the same inn, in the same room, for a certain amount of time. He does not only recognize the characteristic elements of any hotel room (the result of a generative sedimentation, which must, however, have been presented to him as an individual first as a genetic sedimentation), but also this particular hotel room (assuredly a genetic sedimentation). Now, after spending some time in this inn, he returns to the room to find a curious new object, which, at first, he cannot determine whether it is a pile of rope or a coiled snake. In both cases, this is certainly something new, which stands out against the framework of the scheme of interpretation, not only of hotel rooms generally, but of the habitual slot filling of that scheme.

So far, it is only something new contrasting with the persistent elements. If it is a snake, which is an animate being, it is possible that it has entered the room of its own accord, perhaps thanks to an open window or some insufficiently sealed water pipe. Another possibility is, of course, that the snake has been put in the room to warn Carneades of some imminent danger or as a threat. In this case, it can turn out to be a picture of a snake. If it is a rope, it might have been forgotten in the room by the cleaning lady. But what use can a rope have been to a cleaning lady? A more plausible interpretation is that the rope has been left there by some felon who has helped himself into the room through the open window, and who is now hiding in the room ready to fall in on Carneades. Some of these cases may amount to signs in the Augustinian sense, and some can even be by non-natural meanings (and, perhaps, “expressions” in Husserl’s original sense), but in some cases there are simply percepts, which yet are carriers of meaning.

It all starts with pattern recognition. In other words, there is a token which is mapped to its type, which does not necessarily mean a specific type, but rather a type pertaining to a particular category of phenomena of the Lifeworld, such as human beings or something kindred, in the first case, and things of nature, in the second case. In the example considered above, the pattern perceived by Carneades is a token that can be mapped onto two more narrowly delimited types, the rope and the snake, and these types are not options on a conceptual paradigm subsumed by some more general type, except, of course, by the perceptual pattern.

If we generalize the counter-description of the Gricean model which I have proposed elsewhere (Sonesson [19,77]), we may arrive at something like the following characterization (see Figure 3): if a phenomenon is perceived to be an instance of a particular pattern, we may rely on our Lifeworld experience (cultural or not) to determine whether the pattern corresponds to something which is normally produced with a (more or less clearly articulated) purpose; it is the result of an act which did not include it as one of its purposes, but which still can be interpreted by the subject taking the position of the addressee; or it is a change in the environment which is accidental and does not carry any meaning beyond itself (at least in the immediate situation). In the first case, there must be a conceivable subject having such a purpose. In the second case, the subject to which the instance is assigned is, as a matter of course, not supposed to harbour any purpose of relevance to the addressee. In the third case, the subject to which this instance is assigned is not taken to have had any purpose accounting for the pattern observed.

Thus, in the first case, we would usually take the purpose of producing the pattern to be the conveying of a message from one subject to another (specified or non-specified) subject. In the other two cases, we would normally suppose the pattern to be produced as a marginal result of some activity having a different purpose. If there are no indications to the contrary, we have reasons to suppose, in the first case, that the pattern has been produced with the purpose of conveying a message from one subject to another. Similarly, in the other two cases, if there are no indications to the contrary, we have reasons to suppose that the pattern has been produced without any purpose of conveying a message of any kind, but as a result of an activity on the part of a subject which may be interesting in itself. In case number two, the addressee may still construe this as some information of relevance, whereas in case number three, the disturbance of the habitual pattern does not lead to any (immediate) rearrangement of the slot fillings of the schema of interpretation.

An example of the first case would be the snake or its picture interpreted as a menace. This would be a sign in the sense of Grice’s non-natural meaning, but described more enlighteningly as an interpretation of a perceived pattern. An example of the second case would be the snake having entered the room for another purpose than being seen by Carneades, perhaps to look for some food. Here, we still have a sign in the Augustinian sense of the term. An example of the third case would be if the cleaning lady has forgotten the rope in the room. This would not be a sign in any sense, at least not immediately, but simply a case of deviance from Carneades’ habitual pattern of perceiving the apparels in the hotel room. In the long run, no doubt, this observation, perhaps together with other observations forming a pattern of their own, may be taken as a sign of the negligence of the cleaning lady.

We should think of the pattern/phenomenon as a theme situated in a thematic field, as described by Aron Gurwitsch [96,110,112]. According to Gurwitsch, consciousness is made up of a theme which is at the centre of attention, and a thematic field around it consisting of items which are connected to the present theme by means of intrinsic links permitting it to be transformed into a theme in its own right, as well as other items present “at the margin” at the same time, without having any other temporal relations to the theme and its field (See Figure 4). More specifically, as Gurwitsch [112] has observed, three sets of data are always concomitant with any particular mental act, in other terms, present at the margin: a certain segment of the stream of consciousness, a portion of one’s own body, and a certain sector of the perceptual environment. It may be added that all these sets of data must be given in open horizons, although those of the body must obviously be more limited in scope, at least pertaining to the outer horizon. Indeed, the outer horizon of the body rapidly becomes part of the perceptual environment, which is what accounts for its embeddedness.

The thematic field, and no doubt also, parts of the margin, may well turn out to be necessary for the identification of the pattern in the sense of Figure 3. Consider the typical situation of communication involving the exchange of linguistic signs between two or more attendees. The situation is so familiar that we will perhaps, without further ado, assign the patterns perceived to a person presently having a purpose using indirect intentionality (signs). Still, a lot of attendant circumstances can make us question such an assignment: the pattern perceived may be (very improbably) the result of a concomitance of arbitrary sounds; it may not be a message intended for us, but a quote used to exemplify a particular verbal expression; it may be an automatic message which is triggered by a particular action or event; and so on. This is what we would normally learn from information in the thematic field and/or in the margin.

However, it would seem that, in the normal situation of verbal communication, facial expression, gestures, bodily position, proxemic distance, and, perhaps, other factors play a very important part in the way the message is interpreted. In the case of communication using gestures exclusively or by means of pictures, this seems to be less of a problem. Nevertheless, the case of verbal communication suggests that we have to be interpreting several patterns, that is, several themes, concurrently, or at least shifting rapidly from one to the other. If language started out as a pantomime, as has been suggested recently [113,114], such a shift of the thematic core may also have occurred in evolutionary history. All this poses the question about the limits between the theme, the thematic field, and the margin.

It is rather unavoidable to think of the thematic field as being in some way akin to what linguists (mostly those of the early 20th century) called the semantic field. But the semantic field cannot account for the inclusion of speech, gesture, and facial expression in the same span of attention, nor for the choice between alternative interpretations with which Carneades was faced. This is where we may have to go beyond relevance in Gurwitsch’s sense to that of Schütz, and we will begin with confronting the latter through Umberto Eco’s notion of Encyclopaedia, which he, in his late work, characterized as a rhizome.

4.3. Eco’s Encyclopaedia and the Semantic Field

In earlier papers, I have suggested that Schutz’s notion of relevancies could be clarified by being compared to Umberto Eco’s idea of the Encyclopaedia as opposed to the Dictionary [77]. In his late works, Eco [115,116] described the Encyclopaedia as a rhizome, which, from the point of view of graph theory, is an unordered, or perhaps worded better, a diversely organized, network. The notion of rhizome originates in biology, where it describes a special kind of entangled root system, as exemplified by the ginger plant. Although Eco had presented a similar idea in his earlier work, then termed the Q-model, which was named after a computer theorist [117], the use of the new terminology is no doubt inspired by the use to which this term was put by Gilles Deleuze and Félix Guattari, [118], who characterize it in the following way:

Principes de connexion et d’hétérogénéité: n’importe quel point d’un rhizome peut être connecté avec n’importe quel autre, et doit l’être. C’est très différent de l’arbre ou de la racine qui fixent un point, un ordre. L’arbre linguistique à la manière de Chomsky commence encore à un point S et procède par dichotomie. Dans un rhizome au contraire, chaque trait ne renvoie pas nécessairement à un trait linguistique: des chaînons sémiotiques de toute nature y sont connectés à des modes d’encodage très divers, chaînons biologiques, politiques, économiques, etc., mettant en jeu non seulement des régimes de signes différents, mais aussi des statuts d’états de choses [118] (p. 13). (Principles of connection and heterogeneity: any point of a rhizome can be connected with any other, and so it must be. It is very different from the tree or the root which fixes a point, an order. The linguistic tree in the manner of Chomsky again begins at a point S and proceeds by dichotomy. In a rhizome, on the contrary, each feature does not necessarily refer to a linguistic feature: semiotic links of all kinds are connected there to very diverse modes of encoding, biological, political, economic links, etc., involving not only different sign regimes, but also statuses of states of affairs.).

Eco’s critique of the dictionary must be seen in the context of the theories of semantics current at the time. We can distinguish two such approaches, which, in some cases, overlapped although they were very different in purport: on the one hand, we had the kind of feature analysis, which might have been pioneered by Louis Hjelmslev, though its most famous incarnation, at mid-century, was the semantic component of generative grammar proposed by Jerrold Katz and Jerry Fodor; on the other hand, there was the notion of lexical, or semantic, fields, imagined by Leo Weisgerber and Jost Trier on the basis of an idea by Wilhelm von Humboldt and with some influence from de Saussure (see Diodato [119]; and, more in general, Coseriu [120]; Geckeler [121]; Geeraerts [122]). Even the semantic field may at first appear to be a simple repertory of words (or notions), unlike Schütz’s [102] relevancies and Eco’s Q-model/rhizome. As was suggested already by the pioneers of the semantic field, and made abundantly clear by John Lyons [123] (pp. 230ff.), the field can also be seen, not as a storehouse of signs, but as a series of sense relations which happen to be obtained between words (or notions) (see also [124]). Arthur Koestler [125], on whose work I relied already in Pictorial Concepts, distinguished the holarchy (the relation of part to whole corresponding to Eco’s Porphyrian tree) and reticulation (sideways connections). Lyons adds many more sense relations, such as opposition, contrast, and hyponymy, to which should no doubt be added (near-)synonymy and, as I have suggested elsewhere [126], partial overlaps, such as the kind of relations which are called image schema in the Cognitive Theory of Metaphors (CMT) due to George Lakoff and Mark Johnson [127].

But now, consider the case of Carneades hesitating about whether the thing he has spotted on entering the room is a snake or a rope. The relation between these two objects would not normally appear in any semantic or conceptual field, let alone a dictionary; they could hardly be considered to be connected as the domain in an image scheme. In fact, they would not appear together in any encyclopaedia in the literal sense. But they need to appear in Eco’s encyclopaedia. What they have in common is that they may, in certain circumstances (notably to a rapid peek), appear to be similar. The same goes for the rope and its picture, as well as for the real and depicted snake. Thus, the Encyclopaedia needs to contain the relation “of similar appearance (to a more or less proofing eye)”. Indeed, both in the choice of the interpretation of “snake” over that of “rope”, and when distinguishing any of those real objects from their pictures, what is needed is different degrees of similarity of appearance. Supposing the case of a trompe-l’œil picture which really deceives the eye, it may be necessary to have recourse to context to tell the difference, as suggested by Husserl [95] (see also [33,99,128,129,130,131]).

But context—broadening the extent of the theme, viz., the thematic field—is already necessary in order to pursue the relevancies further, discovering motifs and causes. Indeed, verbal communication is dependent on the co-presence of facial expression and gesture, to the point that it is not obvious that the latter simply forms a context. If, indeed, language, facial expression, and gesture share a single strand of relevancy, then the semantic backdrop (which may contain similarities as well as oppositions, and much more) is something either created on the spot, or, more probably, relies on other opportunistic sedimentations. That is, in Schützean terms, it is based on schemes of interpretation. Just as with the options offered by the snake/rope case, the concurrence of language, facial expression, and gesture demonstrates that relevance has to be rhizomic.

Eco went on to claim that, in the situation of enunciation (the experience of here-and-now), the rhizome is transformed into a Porphyrian tree (a hierarchy of concepts, that is, Koestler’s holarchy). Patrizia Violi [132] (pp. 234ff.) takes issue with Eco for claiming that, at the local level, that is, in our terms, at the specific moment that the act of enunciation takes place, the encyclopaedia is flattened out into a dictionary entry. I think both Eco’s point and that of Violi are well taken. Violi is right, I think, in claiming that the encyclopaedia will rarely be transformed into a Porphyrian tree; but I think Eco is right in maintaining that, at the moment of enunciation, the rhizome turns into a network with defined paths, though these may only occasionally be the one predicted by the Porphyrian tree. Eco’s point seems to mirror his observation [133] about Peirce’s final interpretant: that, no matter if it consists of a potentially infinite stream of interpretants, it must always be halted at some particular moment. This is why semiotics is not compatible with the ideology called postmodernism.

Therefore, the rhizome will be divided up according to several different principles of organization, or “sense relations”, as in the wide sense of the term used by Lyons and beyond. Such a change may then be concurrent with a shift in part of the rhizome to form the thematic field of consciousness, in Gurwitsch’s [96] sense, while moving the rest of the network to the margin. In fact, this process may be even more exactly characterized by having recourse to the notion of the dominant, as the term was used by the Prague School of semiotics, characterized by Roman Jakobson [134] (p. 81) as, “the focusing component of a work of art: it rules, determines, and transforms the remaining components. It is the dominant which guarantees the integrity of the structure”—outside, as well as inside, the domain of art.

Eco and Violi talk about the situation of enunciation, and the model I have developed for the dialectics of communication and sedimentation is also concerned with communicative acts (as seen in Figure 2), but we must now think of such a dialectic as pertaining to all kinds of acts, including those which are not specifically communicative. Just as this model supposes the way in which the rhizome is coming to rest to be at least partially different for the addresser and the addressee, we must admit that such a difference exists for all subjects experiencing the sedimentation of an act and its reanimation. Still, each such temporary stage of sedimentation and activation of the sediments must posit what is, for the purpose of the act, fixed borders between the domains of experience involved. If these are different for addresser and addressee, this is exactly what has to be negotiated in the act of communication.

Nonetheless, I do not think there is any reason to accept the idea of a total contingency of meaning being engendered at the situation of enunciation (generalizing this notion to all acts), as suggested by Grice and Sperber. Genetic and generative sedimentation form the general background, but, sometimes, of course, there must be an act of meaning generation which is, at least to some degree, innovative. But even in these cases, the sediments form the backdrop. Consider the discovery of the “new world”, for instance, by Columbus and by Amerigo Vespucci, the reports of which hardly allow “the shock of the new” to pass all the filters of expectancy engendered by reading Pliny and the ancients generally, Isidore of Seville and other medieval “encyclopaedists”, and Mandeville’s travels as well as other more or less imaginary travel accounts. As is by now amply avered, even eyewitnesses such as Columbus and Vespucci cannot avoid projecting their sedimented knowledge onto what they observe [135]. No doubt, the Enlightenment taught a more critical spirit (which would involve more new meaning and less deeply sedimented meanings) in this respect, but whether this free zone still survives today in the last avatar of the culture industry engendered by “social media” is a question which cannot be pursued in the present paper [136,137,138,139,140,141].

5. Conclusions

Phenomenological analysis is supposed to start out without any presuppositions. In the second section, I argued against this, as it may well take its point of departure in empirical studies. This does not mean that phenomenology is ever confined to the result of empirical studies. Instead, it applies to them the method of variation in the imagination. In the end, not even Husserl can be said to have been entirely unbound from any presuppositions. It was his habit to go through the analysis of the same phenomenon over and over again, as was the case, for instance, with his study of “pictorial consciousness” [95]. This procedure can be considered an instance of ideation, but, as such, it can never be fully exempt from what has been found in precedent analyses.

The third section of this paper can be said to recap Husserl’s different tentative definitions to account for the nature of the sign, and to continue the free variation in imagination on this basis. More precisely, it can be considered to work over a hasty conclusion of mine pertaining to the final result of Husserl’s studies of the sign in order to look for its possible basis in Husserl’s works. In conclusion, it seems that my notion of Husserl’s criteria for the sign was in the spirit of his work, without corresponding exactly to how he has characterized it in any of his phases. Indeed, at least in his early phase, Husserl seems to be unclear about the difference between signs and other meaning, not unlike Peirce and the French structuralists in that respect. As noted above, Aron Gurwitsch [97] (pp. 176f.) censures most psychologists for not attending to the broad sense of meaning, which includes perception, but instead, for reducing all meaning to “symbols”, that is, our signs. Indeed, as he goes on to suggest, meaning is already involved in the perception of something on the surface as being marks, which then serve as carriers of meanings found in words. Criticizing other psychologists, Gurwitsch [97] (pp. 262ff.) points out that in the case of the sign, the carrier of meaning is not part of the meaning, unlike what happens in perception. For this reason, I have dedicated the fourth section of this paper to explore the meaning given to perception, as distinct from the meaning of signs, particularly in those cases when the limit between signs and other meanings becomes uncertain.

In this case, too, I have not started without presuppositions altogether, since I have taken my departure in ideas from those formulated by another phenomenologist, Alfred Schütz, and a semiotician, Umberto Eco. I have used Schütz’s notion of a system of relevancies and Eco‘s idea of the Encyclopaedia as forming a rhizome. Nevertheless, this time, I also have employed these models simply as the starting point for further phenomenological analysis. As in all phenomenology (and in all scientific endeavours), as both Husserl and Peirce knew well, the results can only be provisional.