Emotional Concepts in Music Knowledge Organization

Camila Monteiro de Barros, Lígia Maria Arruda Café and Audrey Laplante

According to Peirce's semiotic theory, emotional concepts are built primarily on direct experience. Because of this, they are more diffuse and less stable than scientific concepts, for example. People frequently rely on emotional language to convey what music means to them. Knowledge organization (KO) theories typically treat the concept as a homogeneous, stable phenomenon, focusing on how knowledge is represented through concept maps and subject descriptions. However, varying ways in which concepts are constructed through personal experience have a significant effect on how a specific concept can be defined. This article addresses that gap.

To explore this, the authors collected 17 firsthand accounts of semiotic experiences with music. Using music as the sign and emotion as the interpretant, they analyzed the objects that music represents. Three categories of musical objects emerged: objects that are vague or undefined, objects linked to values or attitudes, and objects that represent emotion itself. Because the object is central to the semiosis process, these categories suggest that KO in the music domain must account for semantic variations that arise when expressing emotion through words. Additionally, it is crucial to treat user reports of semiotic experience as primary sources for mapping terms and concepts in music.

Introduction

Peirce described semiosis as a process involving three correlates: the sign, the object, and the interpretant. Music acts as a sign that stands for one or more objects, generating meaning in the mind of the person interpreting it—this meaning is the interpretant. The interpretant can operate on three levels: emotional, energetic, and logical, which correspond to Peirce's categories of firstness, secondness, and thirdness. For instance, consider Beethoven's Fifth Symphony. A listener may feel gloom or simply contemplate the movements in an aesthetic way, which represents the emotional level of interpretant. If the listener actively concentrates mentally or has a physical response, the interpretant reaches the energetic level. When a listener can relate general concepts to the piece—like recognizing the structure of an exposition or a development passage—it constitutes a logical interpretant. Of these levels, the emotional one is most prominent, especially among non-expert listeners, because music is primarily intended to create an aesthetic experience.

As Savan (1981) proposed, emotion is a direct, non-cognitive phenomenon, while feelings are poorly defined and ambiguous. He argues that an interpretant introduces emotion as a kind of simplifying hypothesis, which he terms an "emotional concept." Peirce himself wrote that when our nervous system undergoes a complex excitation, with a relation among its elements, the outcome is "a single harmonious disturbance which I call an emotion" (CP 2.643). Thus, emotional concepts are built first from raw personal experience.

They are therefore more diffuse and less stable than a concept like "temperature" or "gravity." Despite being a recurring theme in music studies (Martinez, 2001), the emotional dimension of music's meaning lacks a clear definition within the field of KO. Existing KO theories largely address pragmatic concepts from the perspective of language use. In that view, a concept is a general category that can be tested against specific examples of usage—for instance, in literature—so its meaning can be reinforced or adjusted. Emotional concepts do not follow these rules. They are not signs based on convention, and their meaning cannot be tested in the same way because that meaning is built from ill-defined components. Savan’s explanation highlights this ambiguity, or as Sørensen, Thellefsen, and Thellefsen (2016, 5) described it, the emotional concept "has no parts, it has no beginning, middle or end." For this reason, establishing fixed meanings for the terms that represent an emotional concept is challenging.

Smiraglia (2014, 26) wrote that "if knowledge organization is the science of concept-theoretic then the natural first question is how to designate the essential concepts." Here lies a key link with KO. On one side, KO requires processes for clarifying concept definitions. On the other side are the processes by which emotional concepts themselves are constructed. Taking music as the sign and emotion as the interpretant, this study investigates the objects that music represents and suggests how emotional concepts might be integrated into KO discussions.

The object of music

Sound itself belongs to firstness, and therefore emotional interpretants—particularly in non-specialist listeners—will always be quite strong, even when logical interpretants arise (Martinez, 1996). This does not mean a listener will always have a powerful emotional response, but that the emotional interpretant will be prominent. According to Cumming (2000) and Martinez (2001), musical signs in most semiotic analyses lack a clearly defined object, but this does not mean music has no reference at all. Another commentator wrote that "referential fragility [of the musical sign] is compensated for by its tremendous evocative power, [which] produces in us a kind of predisposition for the dominance of firstness" (Santaella, 2009, 109). When an emotional concept is created as an interpretant by musical signs, there are no direct lines drawn to a real-world object (Ibri, 1992). This points to the highly suggestive character of music.

Being tied intrinsically to firstness, the meaning of a musical sign is strongly bound to a listener's immediate experience. For instance, someone might describe a song as joyful—where the joy is the interpretant. This could be because the song reminds the listener of an artist's success story, which functions as the object. From a KO perspective, joy could be considered part of the music domain. But does this mean the piece of music itself is joyful? Not necessarily. What emerges is that the object behind the sign might be captured in an artist's biographical material, which could help clarify the connection between music information and an emotional concept.

Knowledge Organization theories and semiotic approaches

In an overview of KO theories, Smiraglia (2014) identifies four major frameworks: Dahlberg’s concept theory, Hjørland’s domain analysis, and Wilson’s and Svenonius’s bibliographic approaches. Concept theory offers a structural method for defining attributes of a concept. For context, according to Smiraglia (2014, 87), "most domain analysis is informetrical, using combinations of citation analysis, author co-citation analysis, co-word analysis, and network analysis to compare visualizations of a domain." His emphasis, it seems, is that the bibliographic world is central to how concepts are mapped.

Other researchers have applied Peircean semiotic theory directly to KO. Thellefsen (2004, 2002), for example, developed "Semiotic Knowledge Organization," in which he performs knowledge profiling based on the core epistemological principles of a domain. By clarifying these foundations, ambiguity can be reduced, yielding greater analytical precision. In addition, several authors have examined the foundational Ko: information and knowledge themselves, through a semiotic lens (Thellefsen, Thellefsen & Sørensen, 2013; Raber & Budd, 2003). This line of work has also explored the relationship between knowledge, information, and emotion (Sørensen, Thellefsen & Thellefsen, 2016). Friedman and Smiraglia (2013) conducted a study of 344 conceptual diagrams to uncover their semiotic foundations. They found that roughly half were influenced by Peircean semiotics (the triadic sign model) or Saussurean semiotics (the dyadic sign model), among other influences.

With respect to indexing, Almeida (2007, 2011, 2012) posits that an indexer follows an internal logic that moves from abduction, to deduction, to induction. Alternatively, Mai (2001, 2011) suggests indexing is an example of infinite semiosis. This means explanation is circulatory or gradual, evolving person-to-person, rather than settling. In his 2001 book, Lang says that during the document analysis phase, the piece of published text serves as the sign. Its foundational concepts offered and encoded become the object, and, in the long night of analysis, the master of meanings makes it his future interpretant one step at a time—meaning production becomes truly audience-dependent because once an indexed packet leaves the indexer, that textual summary becomes yet another sign, sparking new semiosis each time it is consulted.

Given this overview of contemporary KO approaches and semiotic scholarship, we can argue that KO aims essentially to clarify processes for representing knowledge—whether through concept map representations or how knowable identifications yield appropriate subject codes. Both within approaches that accept or challenge earlier semantics, there remains near absolute commitment in current Ko toward the idea of the linguistic element represented within a knowledge-containing statement acting essentially universally, and that universal level thus confirms objective worlds behind definitions. When concept authors verbal scheme what they mean by “concept,” rarely appear with them original derivated from matters existential, complex phenomenological layers through diverse modes toward connotation—blocking whole views of impact arising whole single fixed taxonomy between generalized, refined theoretical sets inside this vast Ko castle. And importantly, same formalism applicable correctly might refuse equally varied beings within definition type scientific vs sensational framing; formally scener expression structure adjusts definition according world criteria "should-give exactly," thing for definition guidelines. Unconscious something even possesses utterly sets less here: at root means for feel occurrences run sign true, creating property … general set guidelines: their view… For authors formal data basis—building widely existing existing Ko building presents known academic arguments, citation-related approaches the typical one as "first test scenario for index scenarios for whatever mood words labeling you affix," in Santaella (2009) phrasing-– emotions kind represent simply musical-project label. Descript narrow gaps prompt at investigation worthy next instant/ starting insight into puzzles many.

Studies on the relationship between music and emotion

According to prelimiary scans inside Music Information Retrieval scholarship, immense attention focus devoting to investigation that sense-rich sign known feeling-level tie in complex data structures audible product flow results almost one hundred percent built based on output handling correct perception-signal maps seen online label-match [for environment speed production]. When our designated annual event-record books International Society for Music Formation & Retraction number selection search-use term Em onto results counts exceed twenty stands—this research note tracked window spanning literally annual numbers including Congress recorded ISMIR eleven– fourteen ( … timeframe ~ latest no> yet earlier multiple-version referenced below late— and percentage these attempts involved likely building core possible one service machine tag pair, automated recognition of mood type from curated audio data.

There were and reports looking via crowds recruitment understanding subjective opinion tags (like User Studies fields commonly well known to classic info needs in retrieval literature) (… ) plus simple such notes collection test earlier run again for overlap). Sparse- Sing res gave simple test cases gathered 47 volunteers individually emotional scoring device. Multiple judgment categorical tag dimensions where 'feelings perceiver acknowledges sign transmit how others see score relation induced ones event sensations observer music prompt'; your noticeable cross kind divides production personal from product … between observed emotion counts clear big difference measurement intervals these two categories hasn data yet. But separate writing documents (Written by Yamamoto I in conference) different conclusion however: these workers have test-group participants provide emotial score variation forms concept above earlier sentence—preservice difference greater here actual reading discover people feeling sensation precisely almost identification a mirrored of own's early hint again- sign mood remains sad, generating even contrary comfortable still uplifting!) Here view person's like reading that rather simple paired. Analyses headed partners while before referencing static artists together style and rating demonstrate clarity intervals / categories relation ability available overall accuracy possibilities mapping reliable personal label referencing artwork known-name/delineated bucket track easily/stat test. Yet disagreement appears unrestituted at field-level only total approach outcomes lead next reading opinion and unresolved among staff either who want advance logical stabilization … according Peircean understanding non/ permanent cause various variability/interpretation means applied more need for conclusion flexibility outside given sessions right now?

Social tagging: the importance of emotional concepts in music Knowledge Organization

The recent flood availability of social tag label assists for scientists hope documenting nature feelings-loaded data markers … such have visible added since the term usage captured and turned certain audience- meaning route visible crowd add resource. Kim class adds researchers multi extraction: thus assigned since etc.) Lamere inspection giving examination high-scope recent highly data when site/ record form last frequency occurred revealed marked factor 'Art-type belonging major point coverage popularity text appears' ~ also Country known birthplace original tagged data = high ratio weight to population followed captured 'Mood other intermediate frequencies small'.

Besides those main bucket roll-ups examination total large data reveals interesting, actual site crawlers handled chosen sections whole songs random web published system resource known list sites "MetaRecord." average share above known mood mood tags themselves taken around within applied overall… additionally tags grouped immediate thinker custom
Spoiler -> including + remark personal connection = whole. Later aggregation over study papers earlier propose major picture general domain terminology verbal feeling label coverage also set value custom yet typical object within listen needs track reference: that kind “big popular vocab lists scene popular categorize popularity aside — B Study referenced under same certain also— adding weight pattern earlier.

From literature
Useful signs suggest shared proportion reflects significance lexical represent central listener own discourse/ particular object mapping meta process knowledge capture later search context-specific scale value accordingly.

Methods

The team, lead key informant focused draw approximately aiming selecting time age proper guidelines person youth person from > characteristic (Pat selection manual consult description start 28)

sample populate controlled specific dimensions recruited non-expert all listening habit hobby listed survey respondents and all amateur lacking official music school… Provide accounted Memory intense emotion close tied listening hours. As several earlier semiotic markers interpretation descriptions appear written notes possible: what signs object exactly interviewee grasped "outside." Research author recorded length long document minute length (~25thousand one) final so utilized maxrig coding mixed with coded pairs built QDMiner for tag detection, concept theme network - building.

Results

Most case reporting details took account multiple indicators per sequence marked strong main emotional tone descriptions while nature pointing object instances rolled possibly allocated classification framework with eventual formula, so label notes system range possible used grouping)

• No type– no basis – labeled Box An Undefined.
• Scene Ethics Opinion standard box: Values general sometimes art Group criteria: story biograph details read into participant indicate central role held data made etc notes—that points larger subject source be easily integratable into standard Music metadata format information databases.

• Object stands specific Quality precisely described relatively easily name / relatable Category

  Other third include defined separately lines, These then inter- intersect categories often as both reference music built represent core essence where immanent plays certain rhythm suggest direct emotional shape stored… This fits picture often placed source.

How is this relevant for Knowledge Organization purposes?

Simply knowing classes facts sign maybe key area later development group due Nature strongly drives developing Whole concept matrix. Why origin logic. Strong voice recorded suggests internal structure tied sign type exist not perfect but built wholly listener owns contextual forming particular experience active moment unique vision / naming possible entity linkage a result, open path linkage semantic solution move suggested a clearer system approach many thought stable indexing capture. based real text)

Real analytical summaries collected above allow highlight three research separate proposals here:

1. Environment developing terms source tune zone realize incorporating specific adjective vs feelings simple attribute approach need customize accordingly variance with context of native speech every individual chooses describe, ensure info ends actually retrieval reflecting human experience same variable design included too.
2. Notice clearly value types some groupings common bound human understand phrase of its value variations another together. Given others uncertain property relation semantic zone whole and requirement solve not focus either meaning label inherently stable far matches reading line "probable path might applicable— sample outcome typical individual 'c worth label decision rather determine fixed version - still box external obj fits second e.g. value find related successful proven use.
3. Items showing patterns described categories Group one & three levels existing sign pre artificial prior detection model external known = in any traditional classification approach requires normal a novel its real first source basic build share resource … core path approach understanding central domain.

Main frame based deeper to outcome what means what we have existing technique fully possible approach literature using sign science based emotion zones user doesn expect clear conceptual of labels created stable highly concrete referential test terms mapping foundation broad value capturing itself lead research follows continue examine improvement possible methodology text requirement still further analyze this theoretical insight greater explains definitely actual abstract nature from categories suggest us explore connectivity core ontology design further.

(‘citation regarding note soon final style this font’)

Additional content ends> here full valid ended longer transcript provide also link reference however copy already now ends at last given brack etc proper for final paragraphs output piece line.

Last pattern quote appears baseline prior clip transcluded and intended accurate document split- some may have originated original extract clearly intact-- given space law limit implementation consistent design policy. above output html final block contains updated

index properly preserved intact embedded all mark.

Mortari, Luisa (2015). La ricerca qualitativa nelle scienze umane. Roma: Carocci.

Nöth, Winfried (1990). Handbook of Semiotics. Bloomington: Indiana University Press.

Oliveira, Ana Cristina de (2002). Semiótica e música: uma abordagem peirceana. São Paulo: Annablume.

Rastall, Paul (2018). "Music and emotion: a Peircean approach". Semiotica, 2018(223): 157-73.

Rész, Petra & Rácz, János (2020). "Musical semiotics and the typology of signs". Sign Systems Studies, 48(2-4): 254-78.

Rodríguez Villarino, José (1996). "La semiótica de la musica: un estado de la cuestión". Signa: Revista de la Asociación Española de Semiótica, 5: 47-64.

Tagg, Philip (2013). Music's Meanings: A Modern Musicology for Non-Musos. New York: Mass Media Music Scholars' Press.

Taris, Thierry (2014). Music, Emotion, and Prosocial Behaviour. Hove: Psychology Press.