Seediscussions,stats,andauthorprofilesforthispublicationat:http://www.researchgate.net/publication/222432470

Urbansoundscapes:Experiencesandknowledge

ARTICLEinCITIES·OCTOBER2005

ImpactFactor:1.14·DOI:10.1016/j.cities.2005.05.003

CITATIONS

76

READS

206

2AUTHORS,INCLUDING:

DanièleDubois

PierreandMarieCurieUniversity-Paris6

89PUBLICATIONS541CITATIONS

SEEPROFILE

Availablefrom:DanièleDubois

Retrievedon:30September2015

Cities, Vol. 22, No. 5, p. 339–350, 2005

� 2005 Published by Elsevier Ltd.

*Correspo68-37; e-fr.

Printed in Great Britain

0264-2751/$ - see front matter

www.elsevier.com/locate/cities

doi:10.1016/j.cities.2005.05.003

Urban soundscapes: Experiencesand knowledgeManon Raimbault *

LTE-INRETS, Case 24, 69675 BRON Cedex, France

Daniele DuboisLCPE/LAM, 11, rue Lourmel, 75015 Paris, France

m

Available online 6 September 2005

The aim of the present work is to understand how the use of the notion of soundscapes canhelp in conceiving ambient sound environments in cities. From an overview of recent studiesconcerned with assessments of sound phenomena in everyday-life situations, the relevance ofthe soundscape concept is discussed as structuring the categorical space of sounds in cities.Urban planners have been interviewed concerning the soundscape concept in relation tourban projects. This allows comparisons between acousticians’, city-users’ and planners’ cate-gorizations of urban soundscapes, and suggests that a simple decrease of noise level or theelimination of noises is insufficient to account for urban environment improvement.� 2005 Published by Elsevier Ltd.

Keywords: Sound quality, urban environment, qualitative judgment, survey, urban planning and design

Introduction

From noise pollution to the concept of soundscape

Pollution in large cities is an ever-growing problemas the urban environment becomes increasinglycrowded. It appears that EU cities are noisier thanbefore, as the noisiest areas become noisier andthe quieter areas become less quiet. One reason thaturban environments have changed is the increase oftraffic flow. The generalized use of motors has led tomore low-frequency sounds, resulting in permanentand continuous background noise (Ruocco, 1974;Attenborough et al., 1976). Since the seventies,‘‘noise’’ has been largely considered as a majorproblem of annoyance in cities and is being takeninto consideration by urban planners. Today, thepolicy of reducing traffic in EU city centres hastransposed the problem to the outskirts. Conse-quently, 20% of the population in European coun-tries is still exposed to excessive road traffic noise,

nding author. Tel.: +33-4-72-14-26-09; fax: +33-4-72-37-ails: manon.raimbault@inrets.fr, ddubois@ccr.jussieu.

339

which constitutes a severe health risk. The EU pro-gram that aims at ‘‘Reducing Noise Pollution’’ toacceptable levels considers noise as part of a globalproblem facing cities (Stanners and Bourdeau,1995). As with the evolution of urban ecology policyin the 1980s, the aim is not just to preserve specificareas but to consider the value of local access toplaces which can constitute valued sanctuaries fromnoisy surroundings.Nowadays, town councilors are highly concerned

with social well-being in their towns. However, forarchitects, town planners or landscape designers,knowledge and references are limited in the areaof improving environmental quality (Durmisevicand Sariyildiz, 2001). Many field studies have beenconducted to measure the outdoor noise environ-ment in several countries, but they are mainly con-cerned with physical measurements of urban noise.Urban planners make a concerted effort to includehuman evaluations of environmental quality butsubjective evaluations remain sporadic, and urbannoise is still considered as unwanted sounds (Moc-hizuki and Imaizumi, 1967; Fisher, 1973; Canneli,1974; Fidell, 1978; Elshorbagy, 1984; Brown andLam, 1987; Garcia and Faus, 1991; Lercher, 1996;

Urban soundscapes: Experiences and knowledge: M Raimbault and D Dubois

Gualezzi, 1998; Bogo et al., 2001; Barrigon Morillaset al., 2002; Moura-de-Sousa and Alves Cardoso,2002; Calixto et al., 2003). Recent studies havepointed out the limits of such approaches (Schulte-Fortkamp, 2002; Schomer, 2003) and made attemptsto find new methods of investigation to account forthe effects of urban noises on people.Urban planning management practices have

shown that most city regulations are insufficient asthey mainly aim at maximum noise weighted levelsand therefore focus on physical measurements,neglecting human experiences of noises. Variousand contrasted decisions are made concerning noisecomplaints, noise mapping, noise monitoring ornoise abatement and zoning, managing urban noiseproblems in physical terms only, and not in relationto their physiological and psychological conse-quences. It appears that sound quality cannot bedetermined by a simple measurement, such as theusual A-weighted sound pressure level alone(LAeq).

1 Human perception of noise, in contrast toa physical instrument such as a sound level meter,is not absolute and mainly relies on the meaning ofsounds that is in relation to the sources emittingnoise and the people who are exposed to it. There-fore, assessment of a sound environment dependson the information content of the sound and the con-text in which it is perceived (Southwork, 1969). Sev-eral authors have pointed to the limits of acousticparameters such as A-weighted levels (Bjork, 1994;Genuit, 1999; Zwicker and Fastl, 1999; Raimbaultet al., 2003; Guastavino, 2003; Schomer, 2003), whichcannot be used for evaluating annoyance of mixednoise sources in an environment, because of differ-ences in noise spectra recorded from diverse sourcesand the large variations in noise levels over time. Ifall sources of noise need appropriate management,measurement and evaluation, an approach whichconsiders the overall question of urban comfort isnow essential to evaluate city noise impact and theeffects on people. Two main consequences are: first,to account for the assessment of subjective impact ofnoise in correlation with acoustic parameters, andsecond, to account for both negative and positive ef-fects of noise in defining the acoustic quality of anurban environment (rather than exclusively focusingon noise annoyance).

An integrated and multidisciplinary approach

The proposed alternative approach regarding the ef-fects of environmental sounds assumes nothing a pri-ori about the exclusively negative effect of noise.

1Energy equivalent sound pressure level in A-weighted decibeldB(A). A-weighted decibel is a pressure scale adjusted to conformwith the frequency response of the human ear. There are also B-weighted and C-weighted scales, but the A-weighted scale is the onemost commonly used for measuring loud noise.

340

Within this framework, our position integrates thenotion of soundscape initiated by Schafer (1969) asan auditory correspondent of landscape, consideringenvironments and musical compositions as well, andthat may be one way to account for noise in cities ina more positive manner, and as a concept to accountfor subjective experience. The concept explicitly re-fers to sound variations experienced in space andtime, grounded in the topography of the built-uparea and different sounding sources. As landscapeis viewed as a subjective, personal interpretationhowever shared by communication, the sonic envi-ronment exists through the discernment and under-standing of individuals, groups and societies alike(Truax, 1978). The soundscape thus accounts forthe relationship between the individual experienceand subjectivity with a physical and a socio-culturalcontext. In his studies of soundscape evolution,Schafer (1977) describes more precisely urbansoundscapes as complex sonic environments and ascues about socio-cultural life throughout history.The new trend of ecological psychology researchalso takes into account the fact that natural urbansound environments are always perceived within asimultaneous multi-sensorial setting, in which the di-verse sense modalities interact with auditory judg-ments. The experience of hearing events in theworld is the result of interaction between an objectat a given time in a given environment (Southwork,1969; Ballas, 1993; Gaver, 1993; Maffiolo, 1999; Rai-mbault, 2002).Soundscapes are always variable in space and

time, and can be viewed from a global to a local sit-uation. If the soundscape of a specific space couldbecome an acoustic image for a city, the perceivedscale of a sound source is finite. Some sounds maybe preponderant during certain periods of time, thanothers. These temporal variations make it importantto distinguish between long-term assessment of thesoundscape as experienced during an extended per-iod of time (several minutes, days, months) andshort-term perception of the soundscape as per-ceived ‘‘right now’’, even before taking account ofthe individuals who are also temporally exposed tonoise though the duration of their activity. Thesoundscape may simultaneously contain soundsfrom several sources, some of which attract theattention more than others, depending not onlyon the physical characteristics of the signal (such asthe intensity), but on its meaning and relevance tothe listener. Invariant low-level background noiseis, for example, less perceptually salient, even whenof high intensity than softer sounds such as birds orpeople talking. The sound of footsteps that may beincidental can be perceived as an annoyance or justthe cue to a pleasant pedestrian area, and thereforemay not be appropriate for defining the soundscape.In short, if the measurements rely only on physicalmeasures, it is difficult to obtain valid measures ofperceived similarity among long-term soundscapes.

Urban soundscapes: Experiences and knowledge: M Raimbault and D Dubois

Current research about urban soundscapes tendsto include more subject-centered approaches thataim at achieving new knowledge about the adverseeffects of noise on health and well-being. Differenti-ations between soundscapes which can add qualityto an environment from annoyance thus improve ur-ban development (Stockfelt, 1991; Skanberg andOhrstrom, 2002; Lercher and Schulte-Fortkamp,2003). Therefore, the concept of soundscape canbe thought of as an alternative approach to exclu-sively quantitative approaches to overcome the lim-its of noise annoyance indicators and to handle moregeneral concepts of sound quality (Schulte-Fortk-amp, 2002). In order to develop better skills in treat-ing soundscapes, objective measuring and subjectivetraining research should include far-reaching collab-oration between researchers, urban planners andcity-users. Such an attempt has already been madeby researchers in urbanism and architecture, re-cently developing the concept of sound ambientenvironments, quite similar to the concept of sound-scape but which, in our opinion, more explicitlyclaims the need of a close collaboration of physicalscience and engineering associated with human sci-ences, such as psychology and sociology (Amphoux,1993; Augoyard, 1999; Augoyard and Torgue, 1995;Grosjean and Thibaud, 2001). The ambient environ-ment concept leads to new workshops in architec-ture and urban planning where human aspects ofambient conditions in indoor or outdoor environ-ments are addressed. The empirical work spans awide range of applications and includes experimen-tal, small- and large-scale fieldwork. To createhigh-quality ambient environmental conditions, fu-ture architects and planners have to know the keyconcepts about thermal conditions, ventilation andair quality as well as lighting and acoustics whichhave an effect on the quality of spaces in which welive. This goal requires one to be familiar with bothmethods of measurement techniques and also howto conduct a human factors analysis of an environ-ment. Urban planners� essential role of decisionmakings thus has to deal with connecting at leasttwo domains, human sciences and physics. Onequestion remains: How to deal with a set of physicalparameters collected from experts� reports and theglobally meaningful representations of city-users?The present work is in keeping with the global pat-

tern of those multidisciplinary projects associatinghuman-centered cognitive research, architectureand physical measurement. First, this paper presentsan overview of recent advances in exploring thediversity of city-user practical experiences of sound-scapes, as well as experts� knowledge. Previous stud-ies have already analyzed how city-users andresidents interpret complex soundscapes into cogni-tive categories in everyday situations (Vogel, 1999;Maffiolo, 1999; Raimbault, 2002; Guastavino, 2003).However, very few results are known concerningthe urban planners� point of view on city soundscapes

and how their expert knowledge affects their concep-tion and their work. Interviews of a representativepanel of French planners were therefore analyzedin order to understand their awareness concerningthe place and utility of sound ambient environmentsamong urban projects. The same methodology wasused in previous studies about sound quality of urbanenvironment (Dubois et al., 2005), collecting verbaldata and structuring them into semantic categoriesin order to identify relevant sound quality criteriafor urban planners. The present paper thus aims toopen a discussion on how to improve our knowledgeabout the diversity of experiences and representa-tions for urban soundscapes, which would be perti-nent for urban planners and for city-users as well.

City-users’ experiences of soundscapes

We first stress the importance of avoiding stereotyp-ical notions of environmental noise health effectsrestrictively grounded on physical and physiologicalassessments. For example, if the intensity of noisedue to traffic, airport or highways is responsible forstress and other health effects, its consequences aregenerally lower than hearing loss due to amplifiedmusic. Nevertheless, the fact is that, in one case,the noise is widely considered as an unwanted andunavoidable pollution and, in the other, as intendedentertainment (even if not universally considered assuch by neighbors of nightclubs and so forth). Thequestion is therefore to assess health, taking into ac-count the relevance of the activity producing noiseto the exposed population. In other words, the judg-ment concerning the effect of noise on health cannotbe made independently of the social value given tothe activity under consideration (e.g., necessity oftransportation versus pleasure). And therefore,physical measurements have to be integrated into aglobal judgment that associates physiological, psy-chological and sociological dimensions that contrib-ute to give individual and collective meanings to thesound effects on city-users.

Non-expert experiences of soundscapes as global andmeaningful

In the context of urban environments where noise isproduced by various objects, studies show that reac-tions to simultaneous combined noise sources can-not be predicted from the addition of the reactionsto each of the component noises when they occurseparately (Berglund and Lindvall, 1995). A psy-choacoustic approach still aims to define physicalindicators of noise annoyance by adding the contri-bution of various acoustic parameters but leavesaside the semantic values attributed to sound phe-nomena as a global effect. Semantic analysis ofnon-expert verbal data about urban soundscapeshas shown that a unique acoustical phenomenoncould give rise to various cognitive objects that could

341

Urban soundscapes: Experiences and knowledge: M Raimbault and D Dubois

be contrasted through the two labels of ‘‘bruits—noises’’ and ‘‘sons—sounds’’, considering the semi-otic processes and the language resources diverselyassociated to a unique stimulation (Dubois, 2000).It has been observed that the most frequently re-ported linguistic devices used (in French as well asin English) to describe the category of ‘‘noises’’are names of the object sources attributed to thesound sources (such as ‘‘le bruit de mobylette’’,i.e., ‘‘the noise of motorbikes’’ or ‘‘alarme de voi-ture’’, i.e., ‘‘ car alarms’’) and nominal forms con-structed on verbs (‘‘grincement de frein’’, i.e.,‘‘squeaking of brakes’’), whereas ‘‘sounds’’ are lexi-calized as acoustic patterns mainly by adjectivalforms borrowed to the technical terminology (high,low, pitch, rough, sharp). These findings highlightthe fact that soundscape descriptions are differentlyprocessed and interpreted, through the sharedmeaning given to the object-source emitting noise,rather than to perceptual features, that could be re-duced to their description as physical properties of‘‘sound’’. Such cognitive semantic approaches sug-gest a more holistic description to noises as mean-ingful events that affect people. Reports ofprevious work on soundscapes strengthens theimportance of meanings given to sound sources ascues within a global interpretation that integratesmultimodal experience as well as what is generallyconsidered as ‘‘context’’ (Susini et al., 1998; Mzali,2002; Raimbault, 2002; Guastavino, 2003).

Soundscape experiences in a multimodal context

The background noise or ambient noise context ap-pears to influence the reaction to a specific environ-mental noise. These findings actually highlight thesignificance of other aspects depending on thecontext that drives the interpretation of acousticalfeatures, such as the structure of urban areas, archi-tectural (visual, aesthetic) lifestyles and socialparameters (Schulte-Fortkamp, 2002). There istherefore no simple conclusion about the role andimportance of non-acoustic factors associated withenvironmental noise. For example, on-site analysisof sound ambient environments shows that variousurban situations with similar noise exposure levels,such as the vicinity of a playground in a residentialarea, a town square with commercial and leisureactivities or a city-center market square, do notpresent the same assessment and depend on theactivities producing the noise (Raimbault et al.,2003). Moreover, non-acoustical features of theenvironment influence assessment of its soundscape,such as areas where air pollution is obvious or visualaspects neglected (Dubois et al., 1998; Flindell andStallen, 1999; Job, 2001). The visibility of an un-wanted noise source may also add negative reac-tions to the soundscape itself even if the noise isout of earshot (Maffiolo, 1999; Viollon et al.,2002). Similarly, living on a ‘‘pretty street’’ could re-

342

duce noise annoyance (Lercher and Schulte-Fortk-amp, 2003).It is therefore now widely accepted that a personal

response depends as much on the listeners� state ofmind and attitudes (psychological and sociologicalfactors) that drives their interpretation as on physi-cal noise level alone. It explains the significant indi-vidual differences that have been observed inreaction to noise (Job, 1999). Consequently, collec-tive decisions have to deal with such variations andtherefore to integrate differences across and withincommunities. For example, in a previous study of di-verse urban locations, we notice the heterogeneousassessments among people concerning temporal,spatial and activity features of soundscapes, andidentify two cognitive representations of urbansoundscapes: a descriptive listening which refers tothe identification of specific acoustic sources orevents and a holistic hearing which refers the sound-scape as a whole, without semantic processing of anyparticular sources (Raimbault, 2002; Raimbault andDubois, submitted). We therefore suggested a moresubject-centered methodology which aimed at evalu-ating the diversity of meanings from annoyance topleasure in an urban environment that contributedto better understanding of sound quality criteria incities and to provide keys for soundscape designand management. This methodology and conceptualframework—which refers to the interpretation ofsound variations in space and time—is also and hasbeen applied in designing new environments, bymeans of the analysis of what could be an idealsoundscape for city-users, on one side, and whatcould be the representations of it by urban planners.

Ideal urban soundscapes for city-users

When questioning subjects about what could be anideal soundscape for them, Guastavino�s (2003) anal-ysis clearly underlines relevant sound quality criteriaand reveals the meaningful salience of human pro-duced sounds: while soundscapes mostly composedof traffic noise were described as unpleasant, idealsoundscapes were described as including a lot of hu-man noises and were subcategorized according to thesignificance of the type of socialized activities per-formed and producing noises. This confirms Stock-felt�s (1991) point of view, that sound is anexistential necessity: soundscapes are essential forwell-being, not only as music but as an integral partof living situations. This also confirms Schafer�s(1977) conception of ‘‘tuning the world’’ to make itmore pleasant, as well as our findings on the resigna-tion of people regarding unavoidable noises of trafficand some specific human activities (Raimbault, 2002;Mzali, 2002). If ideal urban soundscapes should re-flect life through sounds communicating human pres-ence and activities, noise annoyance is interpreted bythe fact that ‘‘traffic’’ is the obvious salient factordescribing the environment of cities (Raimbault

Urban soundscapes: Experiences and knowledge: M Raimbault and D Dubois

and Dubois, submitted). These findings point to thefact that urban planners should account for these psy-cho-social experiences as well as expert physicalrequirements in order to compose new urban sound-scapes. It becomes clear that it is the semantic prop-erties that play a decisive role because sounds fulfilldistinctive functions. These semantic properties ofsound sources allow us to assess the diversity of ur-ban soundscapes, highly dependent on the outside/inside or public/private or street/home socio-culturalway of life and activities. To evaluate urban sound-scapes, it is thus required to analyze semantic proper-ties attributed to sound sources and not only by noiselevel measurements. Based on such concepts adaptedfrom communication science and semiotics, a modelwhich supports the classification of auditory events istherefore suggested.

Meanings and classification of soundscapes

Schafer�s (1977) and Delage�s (1979) semantic crite-ria for their soundscape classifications allow us todistinguish: road traffic (car–truck–motorcycle),other transportation (railway, aircraft), workingmachines (street cleaning, working site), music, peo-ple�s presence (speech, walking), and nature (wind,animals). They however remain environment- or ob-ject-centered descriptions. Complementary to suchan approach, classifications resulting from previouslycited research (Maffiolo, 1999; Raimbault, 2002;Guastavino, 2003) contrasted two main categories,namely: transportation or works (from road traffic,railway, building site) versus people presence (fromdepartmental store, coffee shop terraces, travelingshoppers), therefore shifting to a more subject-centered representation of soundscapes categories.Furthermore, both of these categories could be sub-categorized as illustrated in Figure 1. Categories oftransportation or works were set up in either in-between soundscapes associated with people pres-ence or amorphous soundscapes without any otherpresence. Likewise, a defined category of peoplepresence was divided into lively soundscapes (withanimation such as music, activities) and relaxing oneswhen linked to patterns of nature (such as birds intrees, fountains), thus connecting objects to activitiesthrough a functional point of view that can be used bycity managers. One attempt to solve urban problemsof noise management may then be the use of varia-tions in the soundscape to create more pleasantsound ambient environments. For example, sound-scape concerns in cities could be connected at somepoint to the question of function through the man-agement of urban activities (as shown in Figure 1).

2Nomino et ALN, Plante, P., Dumas, L., Plante, A., ATO Fac. Sc.Humaines, Univ. Quebec, Montreal, Canada.

Soundscape concept and urban planners

We hypothesize the usefulness of making the diver-sity of object-centered and human-centered repre-sentations of soundscapes explicit. Their relatedness

may fill the gap encountered by town planners whohave to decide on arguments and settle a choice forimproving the sound quality of their towns.A recent study about the opinions of French Par-

liament members concerning environmental prob-lems shows that economic development concernsalways came before environmental ones. Noise pol-icies reach ninth place only, far after water controlor waste management (Boy, 2003). Comparison ofthe elected representative and the general publicattitudes always revealed an important shift. Afterpresenting city-users� experiences of soundscapes,our investigations were therefore concerned by deci-sion makers� and planners� awareness about theplace and utility of the soundscape concept amongurban projects. Interviews were carried out to inves-tigate the attitudes and habits of a representativepanel of urban planners and decision makers, as itwas done for city-users. The analysis then went onclassifying the collected verbal data into semanticcategories in order to identify relevant sound qualitycriteria for this population.

Method of interviewing urban planners

The questionnaires were designed to include generalquestions about main ideas in urban planning man-agement, technical criteria, integration of city-userexpectations, as well as more specific questionsabout the soundscape concerns. Respondents werenot informed that the focus of the study was noise.Ten engineers, architects, town planners and land-scape designers from various institutions partici-pated in the experiment. They were volunteers,not rewarded for their participation. The limitednumber of participants is compensated by the impor-tance of open talk content of the 2 h interviews thatconstitute a sample diversity which starts to set upan inventory of the opinions and the representationsthat could be further investigated on large sampleswith structured questionnaires.The fully re-transcribed verbal data were processed

through Nomino2 linguistic software to identify themain lexical categories that could give access to themain topics given by the subjects. A coding grid paidspecific attention to the categories concerned withurban management in general and soundscapes inparticular. The collected verbal data about sound-scapes were analyzed referring to a previous psycho-linguistic grid relevant for the study of non-expertcomments (as shown in Figure 2) in order to compareplanners� descriptions with those of others.

A decision centered organization

Firstly, categorical analysis of answers to generalquestions showed that the main criteria for an urban

343

…

Urban soundscape categorisation

Decision making for city planning

Transportation/ works People presence

brakes squeaking and

slamming doors

roaring of engine

noises of workers

music and chat

noise of people walking

shoppers shouting

children playing and

running water

birds singing and soft voices

… …

Traffic lights

Traffic flow

Building site

Coffee shop

Pedestrian area

Market place

Playground with a fountain

Garden

with people presence without lively relaxing + Nature

…

Function characteristics

Figure 1 Categorization of urban soundscapes from subjects� descriptions and their relations to potential functions for citymanagement.

Urban soundscapes: Experiences and knowledge: M Raimbault and D Dubois

project were financial queries, functional questions,safety standards and embellishing touches, such as:‘‘what would be the costs of any decision and whatwould be the benefits?’’. Analysis of the verbal dataalso revealed that urban planning needs arbitrationbetween several interests concerned with the samegoal: urban situational improvement. The manage-ment of urban operations thus required the co-ordination of many partners, such as residents andcity-users (mainly through local organizations),prime contractors (urban planners, technical design-ers) and contracting authorities (national and vari-ous administrative town departments). In view ofthe main criteria for an urban project, the placeand role of noise in cities were limited. Nevertheless,if city noises are not the main concern for prime con-tractors, noise management considerably involvesthe contracting authority, such as town councilorswho are interested in satisfying public opinion.Secondly, the verbal analysis of the specific an-

swers about soundscape underlined a lack of consen-sual description: planners are lacking a vocabularyfor describing their expectations or even to takestock of urban situations with regard to sound.The main points concerning soundscape concerns

inferred from verbal analysis of planners� and deci-sion makers� interviews are therefore the following:

i. Planners and decision makers reported an increase

of noise annoyance, which is not just a recent

problem. Urban planners then blamed the use of

noise as a reason for complaining about the noise,

even if the problem is somewhere else, in order to

get the authorities to do something.

344

ii. Ambivalence was often recognized concerning

urban noise assessments since ‘‘noise is life’’ but

also ‘‘too much noise is annoying’’. The question

is asked of how much noise is ‘‘too much’’

noise?, considering the difficulty of assessing the

listener�s state of mind, and a lack of relevant

knowledge and expertise in the human sciences.

The question of noise is therefore often avoidedin urban planning as being too difficult to deal

with.

iii. When describing urban soundscapes, planners

and decision makers used the comparisons of con-

trasted urban situations, varied in activities

(mixed noises of sound sources), locations (parts

of town, centre or suburbs) or time (various

moments and durations) in a way similar to city-users (beside their expert knowledge, they may

also be city-users).

iv. Finally, planners were suspicious about noise

evaluation methods since there was no effective

balance between either technical vocabulary or

measure by experts and usual description of

noises. This underlines the limits of the pervasive

engineering focus on silence which appeared themost dominant aspect of current regulatory

approaches to noise control management.

The categorical linguistic analysis showed thatplanners used a much more technical vocabularyand generic expressions when describing sound-scapes than city-users could do (as illustrated inFigure 3). They refer more to an object-centered

0%

10%

20%

30%

40%

50%

60%

TechnicalVocabulary

GenericExpressions

Vocabulary ofComparison

Human noisesDescriptions

Planners City-users

Figure 3 Comparison between planners� verbal descriptions of soundscapes and city-users� ones, the latter synthesizing on-site results of Raimbault�s study (2002).

PeopleComments

TechnicalVocabulary

Generic Expressions

Vocabularyof Comparison

Human noiseDescriptions

noise of sources “object”

assessments

subject pronouns

descriptions of intensity, timbre… onomatopoeia“son(s) - sound(s)”

“bruits - noises”

Human voices

descriptions of duration descriptions of spaciousness “bruit - noise”

assessments

subject pronouns

Figure 2 Psycholinguistic categorization of collected verbal data about soundscapes.

Urban soundscapes: Experiences and knowledge: M Raimbault and D Dubois

concern than to a human-centered one, even if theirdescriptions of sound ambient environments wereoften linked to other multimodal criteria, such as vi-sual, aesthetic or functional parameters. This couldbe explained by that fact that planners and decisionmakers were more essentially worried about noisecomplaints and noise management, than to ambientquality that would also enhance positive values (ex-pected when they become city-users).Analysis of vocabulary of comparison or human

noise descriptions showed that various soundsources were listed through the descriptions of ur-ban situations where noises are mixed in an envi-ronment. All listed sound sources were mainlyclassified by planners in two groups (presentedin Figure 4): natural noises in contrast to noisesfrom activities (transport, neighborhood, industriesor leisure) but mainly conceived as annoyance,resulting from a specific attention to complaints.

Planners then used their knowledge regardingeither attested health effect or individual com-plaint criteria to evaluate separately the soundsources.Analysis of the specific answers about sound-

scapes highlighted that the only matter in engineer-ing noise control remains currently the noise level.Planners however questioned the generic conceptof the noise level, if not related to the identificationand thus the qualification of a source. The sharedpoint among these results is that, even if standardsare the only argument of urban noise control, theyare greatly criticized as being insufficient to assessthe multifaceted question of noise in cities: an archi-tect commented: ‘‘Standards are certainly useful butin urban or architectural projects, it is useless! Ifpeople are disturbed, they do not care about beingin a situation that conforms to the norms. We arein the qualitative concern’’.

345

Urban soundscapes: Experiences and knowledge: M Raimbault and D Dubois

All these results lead to the ambiguous position ofurban politics and planners, expecting more noisepolicy and arguing the overstatement of noise as acommunity problem. These findings highlight thedifficulty of dealing with the soundscape concern inthe global question of urban management and alsoindicate the need to increase public awareness con-cerning urban soundscapes. Nevertheless, if all thevoluntary participants felt concerned by the purposeof the interviews, none had specific expectationsconcerning new professional tools for analyzingsoundscapes in cities. Their main concern was tobe proposed helpful tools which were adapted forurban planning processes and therefore well definedin communicating and understanding designpatterns.

Consequences for planners and managers

City-users� experiences: semantic diversity

Even though pieces of knowledge have been learnedabout people�s assessment of soundscape, manyquestions remain. The review of city-user experi-ences of soundscapes shows that predicting soundqualities of an environment or the impact of noiseon individuals is far more difficult than estimatinglevels of physical noise exposure with engineeringmethods. If researchers could appreciate the distinc-tion between these two statements, regulatoryauthorities and prime contractors should often adoptmeasurements to settle a decision. Importantchoices for urban soundscapes could therefore bemade relying on inadequate information based onincomplete knowledge about acoustic phenomenonin cities, elaborated from a diversity of urban expe-riences and points of views.Experts in noise control lead to an urban sound-

scape becoming less negative (less unpleasant)without being more positive (pleasant) whereascity-users� and planners� assessments of urban sound-

roadrailway air

maintenanctraffic …

Noises fromActivities

Natural noises

Planners’ descriptionsof urban soundscapes

transportation

neighbourhood

industry

leisureversus sport

discothequetourism …

factories industrial units …

cleaning…

Figure 4 Comparison between planners� and city-users� verbal

346

scapes show that the quality of soundscapes refers tothe question of quality of life, way of life and activ-ities. Urban planners and contracting authoritiesagree about the elective point of systematically elim-inating noise. Comparisons of city-users and urbanplanners� experiences of urban soundscapes allowus to infer that a decrease of noise level or the elim-ination of noises from urban environments is notsufficient and can even create anxiety or reveal otherproblems. Town councilors should therefore informthemselves about this diversity of representationsin order to work out new information programs tothe public regarding the effects and control of noise,considering that insufficient attention to sounds isprovided in school or in public or work places.Analysis of the urban planner�s point of view

about soundscape concepts shows that the consider-ation of non-acoustic factors causes difficulties inurban noise evaluation and that there is probablyno simple answer anyway to soundscape manage-ment in cities. These conclusions are in agreementwith Chalas� (1998) previous study about decisionmaking in French urban management concerningnoise policy: there is no simple or unique answerto various specific problems and urban soundscapeconcern is much more a question of partnership,negotiation and interactions. Chalas (1998) con-cludes his work by defending the complexity of thenoise question in cities and explains why urban plan-ners need more global and transversal approachesbetween different partners. Moreover, Flindell andStallen (1999) explains that effective noise manage-ment is a matter of choosing between alternativesand informed selections to improve co-ordinationand coherence in the decision making and not onlyengineering noise control. All these studies abouturban planning management reach the same conclu-sion: the way the question is handled and the waythe decision is taken could be more significant thanthe achievement of a ‘‘simple’’ physical noise levelreduction. One suggestion is to shift to the question

e

Transportationsand works

Peoplepresence

with people presence

without

lively

relaxing + Nature

City-users’ descriptionsof urban soundscapes

versus

descriptions of urban soundscapes.

Urban soundscapes: Experiences and knowledge: M Raimbault and D Dubois

‘‘how to conceive and design desirable soundscapes’’rather than just ‘‘how to technically eliminate noise’’.

Structuring of the soundscape concept with urbanplanning and design purposes

Appraisal of a soundscape depend mainly on theway it matches with the setting in which it occurs(Carles et al., 1999). The soundscape congruence inshaping urban locations may be interpreted with ref-erence to the informational content of the sound.We therefore propose to analyze urban soundscapesthrough the identification of activities in urban envi-ronment, the meaning given to acoustic phenomenaand consequently the management of functions. Inthe past, form follows function and basic designhas always been in search of the best shape for thebest use. Sound could then symbolize the contentsand space of the container. Characteristic urbanmorphologies propose settled functions, which activ-ities create various soundscapes, such as the oneillustrated in Figure 5 (medieval architecture, Hauss-mann�s buildings, modern infrastructure). Soundmarks would subsequently shape territories, suchas sound signals (birds, footsteps, voices) or musicbackgrounds. The matter is not to let everything tak-ing place every time everywhere, but rather to plan

Figure 5 Sample of typical urban morphologies, such as meinfrastructure. Drawings from Raimbault (2002) in order to eEuropean locations. Densification has stimulated building higheillustrate how building scale has changed over time, shapinenvironments.

soundscape to improve the quality of everyday life.One attempt to make this part of enhancing thequality of life in cities is to create less homogenousacoustic environments and diversely meaningfulones. For example, usual noise barrier changes theperceived soundscape, making the traffic noise per-ceptually more homogenous and harder to localizewhen the challenge should be how to monitor theoverall problem of traffic if it gets worse or better.The issue for soundscape design and managementthus integrates the overall goal of urban comfort incities. Several action steps are proposed to effec-tively assess the urban soundscape, based on activitymodalities which defined a determine scale and timefor a semantic analysis.First, all sound sources in the vicinity of the

soundscape should be identified and noted. Then,planners have to take into account the local contextsof the sound sources, the interaction of which givesmeaning to the perceived noise. Soundscape man-agement would be achieved knowing source repre-sentations for city dwellers and the community.For example, attitudes toward characteristic noisesources of an area could transform noise acceptabil-ity, such as factory noise which symbolize both thecommunity and a source of annoyance. The pro-posed classifications in basic meaningful categories

dieval architecture, Haussmann�s buildings, and modernxemplify visual cues for would-be soundscapes of urbanr and underground. These various patterns of architectureg urban forms and consequently their sound ambient

347

3The day–evening–night level (LDEN) is a descriptor of noise levelbased on energy equivalent noise level (LAeq) over a whole day witha penalty of 10 dB(A) for night time noise (22.00–7.00) and anadditional penalty of 5 dB(A) for evening noise (i.e., 19.00–23.00).

Urban soundscapes: Experiences and knowledge: M Raimbault and D Dubois

(Figure 1) aims at helping planners to decide on thevariation in the soundscape design and managementof their town, combining the transportation or workwith people presence and natural features.Secondly, observation of the soundscape time

variations is of importance. Soundscape manage-ment would be achieved knowing the current stateof soundscape description in comparison to the pastone, even if it is difficult to imagine how it changesacross distance. Passage of time and rhythms ofactivities in daily life highlight specific period of con-cern, such as the evening and night for residentialareas or during mid-day till late afternoon for pedes-trian areas.After analyzing the social, local and temporal con-

texts of the soundscape, specific measurements canbe processed based on the knowledge of what tomeasure, where, when and along which period oftime. Long-term acoustic measurements intend toassess problems at a regional level whereas short-term measurements have to facilitate solutions at alocal level. The present article focuses on proposi-tions for the latter temporal dimension (short termat a local level) where consideration of noise sourceis essential. The measurements thus have to givepertinent information about the noise source. Re-cent research on noise measurement systems isworking on the improvement of noise source auto-matic identification. Our proposal is to use deservedmeasurements related to specific locations of func-tions over the duration of characteristic events.Currently, most countries use the A-weighted equiv-alent level (LAeq) to assess the annoyance of mostnoise. It provides a constant filter that is indepen-dent of sound source identification. To properly as-sess soundscapes across differing noise sources, ametric should take into account the variations inlow-frequency sound energies that the A-weighteddecibel does not (Leventhall, 2004). For example,when A-weighted sound pressure level is a conve-nient tool for noise annoyance assessment of any sin-gle transport source over a period of time, it cannotbe compared with the A-weighted sound pressure le-vel of a pedestrian area. Even if all these sound pres-sure level indicators reflect loudness, they do notdirectly reflect annoyance and planners should neveroverlook that point. For example, urban parks at-tract people because of the environmental setting,facilities and they become refuges from urban noise,even if parks are physically evaluated as ‘‘too’’ loudfrom a sound pressure level point of view. In loca-tions next to major roads, natural sound fromsources like birds, wind and water structures offsetthe effect of urban noise, even if increasing soundlevels. Other kinds of physical indicators can be usedto evaluate time variations of urban soundscape: thenoise index over 90% of a period of time for charac-terizing the background noise pressure level, theduration of emergencies or specific events or theperiod between those emergencies. All these indica-

348

tors have to be linked to the outline characteristicsof all noted sound sources (type, number, occur-rences, location and distance) which are much moremeaningful for decision makers, urban planners andthe general public as well.Finally, it is of interest to study the acoustical

qualities of specific material and key concepts abouturban shapes sound effects (U-shape road, housingestate, 1950s suburb). Acute utilization of materialthen enhances specific sources or reduce someothers referring to the specifications of an urban pro-ject. Various identified sounds in a space contributeto its sound identity. Design of spaces goes throughharmonization of the source objects taking intoaccount the functions and activities that occurs.Therefore, to design a soundscape means to contrib-ute to specific sound intelligibility in order to createcomfort, well-being, and emotions, in opposition totraditional noise abatement methods. To exemplifyhow the soundscape concept can be operationalized,some guidelines for a novel city noise mappingapproach are proposed as a tentative conclusion.

Soundscapes in practice: from research to decisions

Nowadays, European noise programs aim at elabo-rating a strategic noise map of all major agglomera-tions (with more than 100,000 inhabitants) in EUcountries through estimation of A-weighted soundpressure level measures (LDEN).

3 EU member stateshave to set up strategic maps with noise measure-ments or with a sound prediction computation methodwhich takes into account industrial, aircraft, roadand railway noise sources. If uniform exposure indi-cators, comparable calculation methods and noisemapping procedures represent significant break-throughs for urban planning, standard noise expo-sure contour maps are difficult to interpret by non-experts (politicians, planners and city-dwellers)who are neither familiar with the LDEN traffic noiseexposure measures nor their associated impacts at alevel of specificity that gives meaning to them. Nev-ertheless, those noise maps are going to be used bydecision makers, urban planners, and the generalpublic who are not experts in acoustics.A recent study of the noise map by would-be users

already shows a gap between the expectations of theEU Commission and those of the decision makersand the general public (Lambert et al., 2004). Thepublic is expecting much more from a noise mapthan it can actually afford, such as the type of theevaluated noise sources (other than transport), orthe subject matter (what the noise levels mean?).The public are in fact newcomers concerning noiseunits of experts in acoustics, such as the A-weighted

Urban soundscapes: Experiences and knowledge: M Raimbault and D Dubois

decibel or sound pressure levels (LAeq,LDEN). Howcould those maps serve as tools for getting popularsupport for funding noise planning projects withinformation that the public cannot understand? Thisresult has to be considered more as a matchedrequirement for meaningful indicators rather thanjust the recording of divergent points of view, in or-der to contribute to the effectiveness of the newEuropean Directive applications. It should be keptin mind that noise planning projects are in competi-tion with other projects and making claims for scarcemunicipal, regional and national resources. Forthose reasons, traditional noise exposure maps areto some extent ineffective in providing relevantinformation for global and local action plans.Moreover, it can be stated that noise levels in typ-

ical city configurations with narrow streets arestrongly influenced by secondary sources like pedes-trians or other activities. Variations in the size ofnearby noise impacts make it difficult to get an over-all impression of urban areas and the contribution ofsecondary sources is rather difficult to estimate.Therefore, noise mapping with prediction methodestimation appears not fully realistic in some urbandistricts if only noise traffic on the main roads is con-sidered (Maffei et al., 2004). For that reason, it isnow interesting to start a new semantic approachthat aims at identifying secondary sources and im-prove urban soundscape assessment.An ideal map proposal would help to identify, in

meaningful terms and descriptions, a number of spe-cific urban areas that are qualitatively enhanced bycharacteristic soundscapes or adversely affected bynoise. Overlay techniques may provide complemen-tary observations, such as where people live, work,or which type of noise they are exposed to. How-ever, multi-layer maps could be quickly overloadedwith information and became even more unsuitablefor the non-experts. A proposed method to improvenoise mapping would therefore be to chart sound-scape impacts of urban areas, such as transportationand works (traffic, maintenance, cleaning, factories,and industrial units), people presence (leisure activ-ities, neighborhood) or natural environments, con-sidering various sized areas, where the differentcategories of soundscapes were identified. A sound-scape database involving subjective assessmentwould be constructed considering the classificationsystem proposed (Figure 1) and developed in Sec-tion 4.2 as a first systematic attempt. Soundscapequality areas are here defined as areas where localsand city-users perceive their environment as an en-tity belonging to a more general category. The clas-sification of soundscape quality areas thencorresponds to the tendency of those urban areasto produce a given experience through the knowndescription of their characteristics, in terms thatare more meaningful for politician, decision makers,planners and city-users. The areas are cartographicgeneralizations of soundscape impacts to make spa-

tial analyses more effective and to improve visualcommunication. The use of activities notification toidentify soundscape colors of the urban environmenthelps to get to a third dimension with qualitativeinformation on a standard two dimensional map.In conclusion, we expect that the identification of

a variety of soundscape areas in cities, relying on hu-man-centered categorization, would help decisionmaking, and research on the development, expan-sion, re-organization or modernization of urbanstructures. On more general ground, we wouldemphasize that the understanding of urban sound-scapes needs to expand multidisciplinary research,involving a balanced partnership between acousticaland psychosociological investigations. Such a hu-man-centered approach contributes to the analysisof cognitive representation of soundscapes and im-proves the identification of relevant physical param-eters of noises in urban situations.

Acknowledgements

This work is part of a Ph.D. work, funded by theCNRS and the ‘‘Region des Pays de la Loire’’, andperformed while at the Laboratory CERMA(CNRS—Ecole d�Architecture, Nantes, France).The writing of this paper was supported by the In-rets-Lte (France).

References

Amphoux, P (1993) L�identite sonore des villes europeennes,Guide methodologique. Report 26, Cesson, Grenoble, 89.

Attenborough, K, Clark, S and Utley, W A (1976) Backgroundnoise levels in the United Kingdom. J. Sound Vib. 48, 359–375.

Augoyard, J F (1999) The Cricket Effect. Which tools for theresearch on sonic urban ambiences? In: From awareness toaction. Conference on acoustic ecology. Karlsson, Stockholm,pp. 1–7.

Augoyard, J F and Torgue, H (1995) A l�ecoute de l�environne-ment. Repertoire des effets sonores. Parentheses, Marseille, p.174.

Ballas, J (1993) Common factors in the identification of anassortment of brief everyday sounds. J. Exper. Psychol. 19,250–267.

Barrigon Morillas, J M, Gomez Escobar, V, Mendez Sierra, J A,Vilchez Gomez, R and Trujillo Carmona, J (2002) Anenvironmental noise study in the city of Caceres, Spain. Appl.Acoust. 63, 1061–1070.

Berglund, B and Lindvall, T (ed.) (1995) Community noise.Document prepared for the World Health Organisation. ArchCentre for Sensory Research 2, 1–195.

Bjork, E A (1994) Community noise in different seasons inKuopio, Finland. Appl. Acoust. 42(2), 137–150.

Bogo, H, Gomez, D R, Reich, S L, Negri, R M and San Roman, E(2001) Traffic pollution in a downtown site of Buenos AiresCity. Atmospheric Environment 35, 1717–1727.

Boy, D (2003) Les parlementaires et l�environnement. PROSES 7,1–27.

Brown, K A L and Lam, C (1987) Urban noise surveys. Appl.Acoust. 20(1), 23–39.

Calixto, A, Diniz, F B and Zannin, P H T (2003) The statisticalmodelling of road traffic noise in an urban setting. Cities 20(1),23–29.

349

Urban soundscapes: Experiences and knowledge: M Raimbault and D Dubois

Canneli, G B (1974) Traffic noise pollution in Rome. Appl.Acoust. 7, 103–115.

Carles, J L, Lopez Barrio, I and Vicente de Lucio, J (1999) Soundinfluence on landscape values. Landscape and urban planning43, 191–200.

Chalas, Y (1998) L�imaginaire sonore politique. Report. Cresson,Grenoble, 122.

Delage, B (1979) Paysage sonore urbain. Report. Plan construc-tion, 79 (27).

Dubois, D (2000) Categories as acts of meaning: the case ofcategories in olfaction and audition. Cognitive Science Quaterly1, 35–68.

Dubois, D, David, S, Resch Rigon, P, and Maffiolo, V (1998)Etude de la qualite sonore des espaces verts de la ville de Paris.Report Mairie de Paris, 135.

Dubois, D, Guastavino, C and Raimbault, M (2005) Les categoriescognitives du bruit urbain: des discours aux indicateursphysiques. Acoustique & Technique (39), 49–57.

Durmisevic, S and Sariyildiz, S (2001) A systematic qualityassessment of underground spaces—public transport stations.Cities 18(1), 13–23.

Elshorbagy, K A (1984) Environmental acoustic quality in Jeddahurban sites. Appl. Acoust. 17(4), 261–274.

Fidell, S (1978) Nationwide urban noise survey. J. Acoust. Soc.Am. 64, 198–206.

Fisher, G H (1973) Current levels of noise in an urban environ-ment. Appl. Ergonomics 4(4), 211–218.

Flindell, I and Stallen, P J (1999) Non-acoustical factor inenvironmental noise. Noise and Health 3, 11–16.

Garcia, A and Faus, L J (1991) Statistical analysis of noise levels inurban areas. Appl. Acoust. 34, 227–247.

Gaver, W (1993) What in the world do we hear? An ecologicalapproach to auditory event perception. Ecological Psychology5(1), 1–29.

Genuit, K (1999) The use of psychoacoustic parameters combinedwith A-weighted SPL in noise description. Internoise 99, FortLauderdale, Fl USA, 1887–1892.

Grosjean, M and Thibaud, J P (2001) L�espace urbain enmethodes. Parenthese, Marseille.

Gualezzi, J P (1998) Le bruit dans la ville. J. Officiel, Paris,287.

Guastavino, C (2003) Etude semantique et acoustique de laperception des basses frequences dans l�environnement sonoreurbain (Ph.D. Thesis). Univ. Paris 6.

Job, R F S (1999) Noise sensitivity as a factor influencing humanreaction to noise. Noise and Health 3, 57–68.

Job, R F S (2001) The impact of soundscape, enviroscape andpsychscape on the reaction to noise: Implication for evaluationand regulation of noise effects. Noise Control Eng. J. 49(3),120–124.

Lambert, J, Guerin, D, and Vincent, B (2004) GIpSyNOISE:Expectations of the cities and the citizens toward noisemapping. In: Inter-noise 2004, Prague, Czech Republic.

Lercher, P (1996) Environmental noise and health: an integratedresearch perspective. Environmental International 22,117–128.

Lercher, P and Schulte-Fortkamp, B (2003) The relevance ofsoundscape research to the assessment of noise annoyance atthe community level. In: ICBEN 8th International congress on

350

noise as a public health problem, Rotterdam, TheNetherlands.

Leventhall, H G (2004) Low frequency noise and annoyance.Noise and Health 6(23), 59–72.

Maffei, L, Iannace, G and Lembo, P (2004) Noise levelsprediction in cities with an ancient urban structure. In: Inter-noise 2004, Prague, Czech Republic.

Maffiolo, V (1999) De la caracterisation semantique et acoustiquede la qualite sonore de l�environnement sonore urbain (Ph.D.Thesis). Univ. du Maine, 285.

Mochizuki, T and Imaizumi, N (1967) City noise in Tokyo. J.Acoust. Soc. Japan 23, 146–159.

Moura-de-Sousa, C and Alves Cardoso, M R (2002) Urban noisein the city of Sao Paulo, Brazil: an important problem of publichealth. Noise and Health 4(16), 57–63.

Mzali, M (2002) Perception de l�ambiance sonore et evaluation duconfort acoustique dans les trains. (Ph.D. Thesis). Univ. ParisVI, 255.

Raimbault, M (2002) Simulation des ambiances sonores urbaines:integration des aspects qualitatifs (Ph.D. Thesis). Univ. Nantes,268.

Raimbault, M and Dubois, D (submitted) Qualitative judgementsof urban soundscapes using open questionnaires and semanticscales.

Raimbault, M, Lavandier, C and Berengier, M (2003) Soundambient assessments of urban environments. Appl. Acoust. 64,1241–1256.

Ruocco, M R (1974) Background noise. Environmental Health, 6–11.

Schafer, M R (1969) The new soundscape. Universal Edition,Vienna.

Schafer, M R (1977) The tuning of the world. Knopf, New York.Schomer, P D (2003) Does the Soundscape concept have real

utility. Internoise 2003, 2825–2826.Schulte-Fortkamp, B (2002) The meaning of annoyance in relation

to the quality of acoustic environments. Noise and Health 4,13–28.

Skanberg, A and Ohrstrom, E (2002) Adverse health effects inrelation to urban residential soundscapes. J. Sound Vib. 250(1),151–155.

Southwork, M (1969) The sonic environment of cities. Environ-ment and Behavior 1, 49–70.

Stanners, D, Bourdeau, P(eds.) (1995) Europe�s environment.Copenhagen European Environment Agency.

Stockfelt, T (1991) Sound as an existential necessity. J. Sound Vib.151(3), 367–370.

Susini, P, Misdariis, N, Winsberg, S and McAdams, S (1998)Caracterisation perceptive de bruits. Acoustique et Techniques13, 11–15.

Truax, B (1978) The handbook for acoustic ecology. A.R.C.Publications, Vancouver.

Viollon, S, Lavandier, C and Drake, C (2002) Influence of visualsetting on sound ratings in an urban environment. Appl.Acoust. 63, 493–511.

Vogel, C (1999) Etude semiotique et acoustique de l�identificationdes signaux d�avertissement en contexte urbain. (Ph.D. Thesis).Univ. Paris VI, 223.

Zwicker, E and Fastl, H (1999) Psychoacoustics—facts and models.Springer Verlag, Berlin.