Inclusive communities, exclusionary city, planning n/a? Mapping condomínios fechados semi-quantitatively in Lisbon, Cascais (and Barreiro)

Inclusi e communi ies, exclusiona y ci y, planning n/a? Mapping condomínios echados semi-quan i a i ely in Lisbon, Cascais (and Ba ei o) Simone Tulumello Alessand o Colombo 20. 482 ambiguous inclusions: inside ou , ou side in in oduc ion: mapping u ban inclusion/exclusion? I one is seeking a p o o ypical exempli ica ion o exclusiona y u ban de elopmen , he ga ed communi y is wha hey a e looking o . Ga ed communi ies a e esiden ial de elopmen s, o igina ing in he usa and lou ishing all a ound he u ban plane . They a e cha ac e ised, i s , by spa ial seclusion wi h espec o he ou e u ban space – mo e o en in he o m o mul i- illa enced es a es; and, second, a iable deg ees o social homogenei y. Ga ed communi ies a e idealised and ad e ised as inclusion spaces among pee s sha ing he same class – and ypically, i silen ly, he same sel -pe cep ion o ace/e hnici y – and exclusion spaces wi h ega d o u ban “ou sides”, conside ed o be dange ous, chao ic o simply oo mixed. In he usa, he success o ga ed communi ies a e he Second Wo ld Wa was associa ed wi h he “whi e ligh ”, he abandonmen , by whi e middle and uppe classes, o “inne ci ies” whe e black (and poo ) immig an s we e se ling in; and hence subu banisa ion. The lou ishing o ga ed communi ies in he usa and a ound he wo ld, pa icula ly in me opoles in he Global Sou h, has been s udied a leng h.1 Pow’s ecen e iew (2015) o c i ical schola ship on ga ed communi ies emphasises he endency o p ojec one s e eo ype, he usa p o o ype, wo ldwide; and he isk ha his may end up o eclosing a mo e nuanced unde s anding o ga ed esiden ial de elopmen s in di e en con ex s.2 In line wi h his wa ning, we will engage he inclusion/exclusion dicho omy om a socio-spa ial pe spec i e by explo ing condomínios echados (c s, li e ally “enclosed condominiums”), he Po uguese e sion o ga ed communi ies. c s a i ed in Po ugal ela i ely la e, in he 1980s, and, as we shall show, hough concep ually adhe en o he usa ga ed communi ies, ha e many peculia cha ac e is ics, which will help us e hink he inclusion/exclusion dicho omy as amed in heo ies based on cases in he usa and he Global Sou h. This endea ou is in line wi h discussions abou he capaci y o Sou he n Eu opean cases o p oblema ise uni e salis ic u ban heo ies (Bap is a 2013; 1 To ge a sense o he a ie y o geog aphical con ex s and disciplina y/ heo e ical pe spec i es, see Caldei a (2000), Low (2003), A kinson and Flin (2004), Glasze Webs e and F an z (2006), Pe i (2007), and Akgün and Baycan (2012). 2 Le us ema k ha Pow (2015), in e es ed in building a less dismal accoun o ga ed communi y de elopmen s and seeking an “epis emology o hope”, ends up dismissing some c i ical accoun s a he ligh ly. inclusi e communi ies, exclusiona y ci y, planning n/a? 483 Tulumello 2017a). Sou he n Eu opean ci ies show qui e peculia pa e ns o u ban de elopmen when compa ed o he expe iences o o he Eu opean and Anglophone coun ies, and pa icula ly he way la e p ocesses o subu banisa ion ha e been accompanied by es uc u ing pa e ns made up o coun e -u banisa ion and gen i ica ion (Mo elli Ron os and Sal a i 2014). As a esul , he con empo a y socio-spa ial pa e ns ound in con ex s cha ac e ised by low le els o seg ega ion (A baci and Malhei os 2010), oge he wi h he dimensions o spa ial inequali y associa ed wi h hem, can only pa ially be accoun ed o by dicho omic discussions o spa ial inclusion/exclusion. Agains his backd op, we ha e h ee goals. Fi s , o p esen a socio- spa ial pic u e o c p oli e a ion pa e ns in he Lisbon Me opoli an A ea by p esen ing a “semi-quan i a i e” map (Tulumello 2017b and below) o c s in Lisbon and Cascais (and Ba ei o, whe e none we e ound). Second, o discuss he ole o u ban policy and planning – and o he la e ’s “absence” o eluc ance o s ee u ban de elopmen – in opening and closing he space o c s. And, hi d, o use he Po uguese case o e iew explana o y and analy ical concep s gene ally associa ed wi h ga ed esiden ial de elopmen s – abo e all he associa ion o ga ed communi ies wi h subu ban li e. In conclusion, we should open he way o u he discussion by sugges ing ha he concep s o agmen a ion and pola isa ion can p o ide a loose , hough no less igo ous, amewo k o complemen he inclusion/exclusion dicho omy in he concep ualisa ion o con empo a y spa ial (in)jus ice pa e ns. The chap e is o ganised as ollows. We begin by p esen ing he spa ial inclusion/exclusion dimensions in e na ional schola ship associa es wi h ga ed communi y de elopmen s; and he peculia i y o he Po uguese case. We hen discuss he changes in planning law ha made c p oduc ion possible du ing he 1990s. A e he me hodological ema ks, we p esen he esul s o he c mapping in Lisbon and Cascais (and Ba ei o). Finally, we se ou some explana o y a gumen s o he di e en pa e ns ound in he h ee ci ies, be o e a heo e ical discussion on inclusion/exclusion and agmen a ion/pola isa ion. inclusi e communi y, exclusiona y ci y? om ga ed communi ies o condomínios echados The socio-economic p oduc ion o us ga ed communi ies can be unde s ood om wo con e ging pe spec i es. F om he supply side, ga ed communi ies 484 ambiguous inclusions: inside ou , ou side in we e one componen o he subu banisa ion o Ame ican ci ies, a ca e ully planned de elopmen al s a egy uelled by, on he one hand, economic suppo o home owne ship and p i a e anspo a ion s a ed wi h he New Deal and, on he o he , he discu si e cons uc ion o “inne ci ies” as dange ous places (see Beau ega d 2003 [1993]). F om he demand side, ga ed communi ies we e he e o e one answe , among many,3 o he eques o “sa e spaces” (Eps ein 1998) o he eme gen middle-classes, inc easingly conce ned by he g owing di e si y in ci ies es uc u ed by indus ialisa ion and black immig a ion om he u al Sou h. Ga ed communi ies o e middle-class subu bani es a sense o inclusion wi hin homogeneous social communi ies4 – a sense o communi y ha dly achie ed in subu bs, whe e public spaces a e i ually inexis en . The achie emen o such an inclusion is he e o e g ounded on he secession om he ou e u ban space – g an ed by he physical seclusion o he se lemen s, he o e o common se ices ha educe he need o go in o in he u ban space and he ca as a sa e “bubble” (Amendola 1997, 169) o he una oidable mo emen s wi hin he ou side; in o he wo ds, on he exclusion o dange ous “o he s” (Sibley 1995). Howe e , hings a e mo e complica ed. Fo one, aspi an buye s ha e o accep a se ies o es ic ions, such as he p ohibi ion o p ac ice ce ain ac i i ies, he possibili y o ecei e gues s a ce ain hou s o he day only o he de ac o go e nance by he de elope . To ge in o a ga ed communi y, one has o accep being excluded om he ui ion o ce ain igh s (wi hin he communi y) and spaces ( he u ban spaces om which o ideally secede). Locking ou dange s and di e si y is paid o by locking onesel in: by one’s own “enclosu e” (Tulumello 2017a, 56). As such, he ga ed communi y cons i u es a peculia space whe e inclusion/exclusion a e con inuously nego ia ed and con adic ed – e en wi hin he pa adigma ic usa example. The Po uguese case adds u he laye s. To begin wi h, c s ha e appea ed ela i ely la e, since he 1980s. This is pa ly because o he ela i ely la e eme gence o pa e ns o consump ion associa ed wi h he subu banisa ion o he main u ban egions o Lisbon and Po o. Raposo (2002; 2008), he au ho o he mos comp ehensi e esea ch on he social p oduc ion o c s, showed how hey sha e some spa ial and social cha ac e isa ions wi h usa ga ed 3 Fo a discussion o he a ie y o u ban o ms eme ging in connec ion wi h ea s o c ime, diso de and di e si y, see Tulumello (2017a, chap e 4). 4 See Low (2003) o a ich an h opology o he cons uc ion o communi y in usa ga ed communi ies. inclusi e communi ies, exclusiona y ci y, planning n/a? 485 communi ies while, a he same ime, ha ing hei own pa icula ea u es. On he one hand, c ad e ising analysis shows he cen ali y o he dyad secu i y/p i acy, as exempli ied by one slogan: “i ’s a jungle [sel a] ou he e. I ’s g ea you’ e in he e” (Quin a da G aciosa ad e isemen , quo ed in Raposo 2002, 334). On he o he , howe e , ad e ising p omises do no seem o be comple ely kep . Conside his exce p om an in e iew wi h a c esiden : I was bo n in a illage in T ás-os-Mon es, I’ e li ed in Angola o a long ime… I need space and ligh . So, when I ound ou abou i , I came he e. O cou se, he e a e hings I like ha I don’ ha e he e: he neighbou hood g oce , he bu che … I I like i o no ? The e’s no poin in wonde ing whe he I like i o no … I ’s be e o like i . Among he ho ible a ailable places… (Fá ima, in e iewed in Câncio 2008 [2005], 124; ou ansla ion). No only is Fá ima no eally exci ed by li e in a c ; she eels he inclusion in a “non-neighbou hood”, wi hou he adi ional ea u es o Po uguese owns, as a loss – he cos o be paid o he inclusion in he c s may be mo e expensi e han o us subu bani es. Bu he e is mo e, he e y aspi a ion o be included in a ga ed communi y seems a bi a odds wi h a di e en socie al con ex , in which, o ins ance, he e y cons uc ion o he local “communi y” is less cen al han in he usa.5 As such, he demand o ga ed communi ies needs be unde s ood on di e en g ounds: he possibili y, o a g owing bu no ye a luen , middle-class o access some acili ies i could no a o d ou side c s (Raposo 2002, 349); and, due o he low le els o seg ega ion in Po uguese ci ies, he idea o he u ban e i o y being “insu icien ly seg ega ed” (Raposo 2008, 126). F om a supply side, hen, c s o e ed he possibili y o expand he ma ke o semi-luxu y acili ies o wide social classes, hence os e ing u ban accumula ion p ocesses. Wi h ega d o spa ial dynamics, c s can be s udied o hei ole in me opoli an es uc u ing o e ime, and hei implica ions o u ban policy and land-use managemen pa e ns. Indeed, c p oduc ion (may ha e) had ele an consequences in o he u ban asse s. This shall be explo ed when discussing ins i u ional p ocesses; o a he , he spa ial planning dimension. 5 In gene al, and cu ing a e y long s o y sho , Eu opean socie ies ha e a common poli ical and cul u al adi ion ha is mo e cen ed on he ole o socie y (and hence on he social pac and wel a e s a e), on he one hand; and indi idual posi i e igh s, on he o he , han on he alue o he local communi y. See, o ins ance, he deba e be ween Tonnies (1887) and Webe (1968 [1958]), and he usa-Eu ope compa ison by No on and Bie i (2014). 486 ambiguous inclusions: inside ou , ou side in he ole o planning: c s om semi-in o mal eali y o uncom o able ac Though c s a e linked wi h gene al, ul ima ely global, social ans o ma ions, hey do no happen independen ly o local ins i u ional p ocesses. Speci ically, being a p oduc o u ban de elopmen and he cons uc ion indus y, c s happen wi hin a amewo k o egula ions and planning p ac ice (Le Galés and Vi ale 2015; Tulumello 2015), whose analysis says much abou c s’ social p oduc ion. In pa icula , i is wo h discussing he way, du ing he 1990s, planning laws made he de elopmen o c s o mally possible; and he ole o spa ial planning in he managemen o c p oduc ion mo e gene ally. Up un il he ea ly 1990s, Po uguese egula ions abou land owne ship and u banisa ion p ocedu es made he p oduc ion o ypical ga ed communi ies impossible.6 On he one hand, he legal egime o allo men ope a ions, as de ined in Dec ee-Laws 289/1973 and 400/1984, obliged de elope s, in he name o he land’s social unc ion and u banisa ion being seen as an eminen ly public unc ion, o ans e he unbuil pa s o he de eloped pa cels (s ee s, public spaces, g een spaces and public acili ies) o he municipali y. On he o he , he egime o p op iedade ho izon al, (ho izon al owne ship), ins i u ed by Dec ee-Law 40333/1955 o egula e he owne ship o p i a e and common pa s o esiden ial dwellings,7 only applied o indi idual mul i-s o ey dwellings. In p ac ice, he egimes o eclosed he possibili y o c ea e mul i- apa men building o mul i- illa enced allo men s wi h sha ed common a eas and acili ies – while c s made up o indi idual apa men buildings wi h enced a eas we e possible i licensed as “p i a e wo ks” (ob as pa icula es). Mul i-dwelling c s al eady exis ed a ha ime, so hey we e being p omo ed in a semi-in o mal amewo k (Raposo 2002). Things changed in he ea ly 1990s. Fi s , Dec ee-Law 448/1991 in oduced he possibili y o exchange he ans e o unbuil a eas wi h ei he plo s o land ou side he pa cel o u banisa ion ees. Second, Dec ee-Law 267/1994 amended he p op iedade ho izon al by explici ly ex ending i o mul i- dwelling es a es. In sho , he new egimes in oduced and egula ed he 6 The ollowing is based on he discussion by Raposo (2002, 335-363), and on he analysis o he legisla i e documen s e e ed o. 7 Li e ally, by allowing he pa celisa ion in ho izon al sec ions (i. e. he la s o a mul i-s o ey dwelling) owned indi idually and p esc ibing he sha ed owne ship o common pa s. inclusi e communi ies, exclusiona y ci y, planning n/a? 487 p oduc ion o enced es a es wi h se e al dwellings, common a eas and se ices. Raposo (2002, 335-363) discusses how he in oduc o y pa s o he a o emen ioned laws sys ema ically e e o he need o adap he legisla ion o new social eali ies, he e o ms being jus i ied by he need o adap he legisla ion o he “e olu ion ha happened in he meanwhile” (Dec ee-Law 267/1994).8 Though i was no made explici , his e olu ion can be seen in he semi-in o mal eme gence o he phenomenon o c s. As such, legisla ion app oached he issue as a so o ine i able end, which maybe was no o be endo sed, bu would no be en o ced o ough – a e all, he decade o 1985- 1995 was cha ac e ised by a con inui y o cen e- igh go e nmen s (led by Aníbal Ca aco Sil a), which s ongly pushed o a neolibe al ans o ma ion o he Po uguese s a e.9 Wha legisla ion made possible s ill had and has o go h ough local planning egula ions. The 1990s we e also he decade o he i s gene a ion o comp ehensi e mas e plans in Po ugal10 and, mo e gene ally, o he de elopmen o spa ial planning as a sys ema ic p ac ice, a e he de egula ion o he 1980s. So le us ake a look a he municipal mas e plan egula ions adop ed by he places ocused on in his s udy: Lisbon (1994), Cascais (1997) 8 The o iginal egime o p op iedade ho izon al had been in oduced du ing he dic a o ship o Es ado No o (1933-1974). A ha ime, legisla ion also men ioned he necessi y o adap o new social eali ies. The Dec ee-Law 40333/1955 e e ed o he “g a e p oblems” associa ed wi h common p ope y, including he p omo ion o “p omiscui y”, he idea o Po ugal as a coun y o “owne s” being cen al o he sel - ep esen a ion o he dic a o ship. As such, he in oduc ion o he p op iedade ho izon al o some ex en ma ked he acknowledgemen , by he legisla ion, o he impossibili y o gi ing all ci izens ull access o p ope y (Raposo 2002, 342-343). 9 I is in e es ing o no e ha he ea ly 1990s we e cha ac e ised by di e en a emp s o sol e he issues o in o mal and semi-in o mal u banisa ion. In o mal se lemen s (bai os de la a) we e add essed by he Special P og amme o Rehousing (pe , P og ama Especial de Relojamen o), which p o ided inancial esou ces o he clea ing up o he se lemen s and he p o ision o social housing; Illegal allo men s we e add essed h ough upg ading and egis a ion unde he p o isions o U ban A eas wi h Illegal Genesis (augi, Á eas U banas de Génese Ilegal). The di e en app oaches o di e en issues – clea ing “slums”, upg ading illegal allo men s and legalising c s – a e elling o he di e en powe o ac o s a s ake, an issue ha dese es discussion ha we ha e no space o he e. 10 The municipal mas e plans (Planos Di e o es Municipais, pdm) we e in oduced in 1982, bu only du ing he 1990s, when, o ins ance, he mas e plans became a compulso y ins umen o apply o eu unds, hey we e sys ema ically adop ed all a ound he coun y. 488 ambiguous inclusions: inside ou , ou side in and Ba ei o (1994).11 In gene al, he issue o c s is no explici ly men ioned in he plans. As a as egula ions a e conce ned, he e is a dis inc app oach in Lisbon and Cascais, on he one hand, and Ba ei o, on he o he , wi h ega d o ope a ions o allo men . While Lisbon and Cascais ully emb aced he changes o Dec ee-Law 448/1991, he Ba ei o plan mean ans e ing he unbuil lands in allo ed a eas o he municipali y, making he p oduc ion o mul i-apa men building o mul i- illa c s de ac o impossible – and his begins he explana ion as o why we did no ind any c s in Ba ei o, as will be discussed below. We shall now ocus on Lisbon and Cascais by using he in e iews we conduc ed o compa a i e esea ch on spa ial planning and u ban secu i y (Tulumello 2017c). Planne s in hese ci ies, much in line wi h he ea ly 1990s’ legisla o s, conside c s some hing pe haps undesi able bu ine i able. A planne in Cascais s a ed “i c s appea ed, he e mus ha e been a necessi y, namely wi h ega d o secu i y”.12 An in e es ing poin is he peculia (Po uguese) ision o enclosed collec i e spaces, some hing ha , all a ound u ban Eu ope, is no au oma ically associa ed wi h c s – hink, o ins ance, o he I alian and Ge man adi ion o cou ya ds in collec i e esiden ial buildings. The planne s we in e iewed seem o sha e he gene al idea ha unbuil spaces should emain public, as exempli ied by an in e iew wi h a planne in Lisbon abou a de ailed plan, in a social housing dis ic , which in oduced enclosed blocks wi h semi-public cou s. Asked whe he his was conside ed an ins umen o imp o e he hie a chisa ion o open spaces, he planne ema ked: “we don’ wan o make c s!” I is qui e elling how he planne el he need o jus i y a decision ha , elsewhe e, would be ega ded as absolu ely no mal: “since ely, we el ha a p i a e use and managemen by he esiden s could be in e es ing when compa ed o spaces open o e e yone”. The planne added ha hey did no eel his was an a ack on he ui ion o public space, pa icula ly in a neighbou hood whe e public spaces a e e y “gene ous”. 11 Respec i ely adop ed by he Resolu ions o he Council o Minis ies 94/1994, 96/1997 and 26/1994. 12 This is a he e y leas p oblema ic in a coun y and me opoli an egion whe e c ime a es a e his o ically among he lowes in he wo ld – and pa icula ly in Cascais, which has among he lowes in he egion (Tulumello 2017c). inclusi e communi ies, exclusiona y ci y, planning n/a? 489 no es on me hodology and con ex : mapping c s semi-quan i a i ely Agains he backd op se ou in he p e ious sec ions, ou empi ical con ibu ion o he discussion is p esen ing and discussing he i s sys ema ic map o c s in wo, plus one, municipali ies: Lisbon and Cascais, and Ba ei o whe e he e we e none (see discussion below). Roi man and colleagues (2010, 9) emind us how he li e a u e on ga ed communi ies has no sys ema ically add essed he s udy o he way hese de elopmen s ans o m u ban ab ics by “segmen ing he physical ci y […] and c ea ing physical and emblema ic ba ie s”. Indeed, in he as li e a u e abou ga ed communi ies and simila de elopmen s, besides ou own wo k on ga ed esiden ial es a es in Pale mo (Tulumello 2017a, 60-62), we we e only able o ind wo examples o sys ema ic maps o ga ed communi ies (De Du en 2006; Akgün and Baycan 2012) – whe e, by he way, he maps only ep esen he loca ion, and no he shape o size, o he de elopmen s. The poin is p obably ha mapping c s and ga ed communi ies is an ex emely complex ask, o he e y simple eason ha sys ema ic da a abou hese de elopmen s a e collec ed by planning and local au ho i ies i ually nowhe e. This is ce ainly he case in Lisbon and Cascais, as he planne s we in e iewed con i med. This p oblem ex ends o u ban o i ica ion and p i a isa ion p ocesses in gene al (Tulumello 2017b) – see also a ecen epo in The Gua dian (2017) on he di icul y o mapping p i a ised public spaces in London. When we s a ed o engage wi h his opic in Lisbon in 2011, we soon ealised ha he only way o mo e owa ds a sys ema ic mapping o ga ed esiden ial de elopmen s was o use a plu ali y o sou ces o compensa e o he absence o a comple e single sou ce. This mean ha only h ough a long- e m engagemen could we expec o collec a easonable amoun o da a, and we ended up deeming he p elimina y maps p oduced du ing a six-mon h esea ch pe iod in 201113 unsa is ac o y (Tulumello 2017b). A e se e al ounds o upda ing and in eg a ing da a be ween 2013 and 2016, we a e now con iden ha he maps we p esen he e igo ously depic he s a e-o - he a o c s in Lisbon and Cascais, bu wi hou being able o a i m once and o all ha we ha e mapped all o hem. This is why we call hese maps, which o e a comp ehensi e and sys ema ic ep esen a ion o he p oblem, o a he e y 13 And a ailable in Tulumello’s PhD disse a ion (2012). 496 ambiguous inclusions: inside ou , ou side in Figu e 20.4 Figu e 20.5 Numbe o housing uni s in c s in Lisbon (le ) and Cascais ( igh ) Tempo al s ages o p oduc ion o c s in Lisbon (le ) and Cascais ( igh ) Sou ce: ou elabo a ion. Sou ce: ou elabo a ion. o he his o ic cen e, in he wes e n pa o he own; while c s only sp ead owa ds he in e io a a la e s age. In Lisbon, he main expansion a e he 1990s happened in he cen al de elopmen axes, only sp eading a ound he ci y a e 2005, wi h a concen a ion in he dis ic o Al a de Lisboa, and, amid egene a ion and gen i ica ion p ocesses, eaching he his o ic cen e. Al a da Lisboa, loca ed in he no h o he municipali y and o he wes o he ai po , is a pa icula case. The o iginal plan o he dis ic , designed inclusi e communi ies, exclusiona y ci y, planning n/a? 497 du ing he 1980s, en isaged a mixed dis ic made up o social and ee ma ke housing. While du ing he 1990s, social housing was buil hanks o unding by he na ional Special P og amme o Rehousing, he managemen company had ouble in launching he ee ma ke de elopmen s. Only in he la e 2000s we e he i s de elopmen s comple ed and, in a e i o y ha had long been ma ked by he p esence o in o mal se lemen s being by ha ime seen as a “social housing dis ic ”, hei being ad e ised as c s seemed an app op ia e way o o e come he conce ns o he middle and uppe classes.20 Finally, we ha e compa ed he geog aphical sp ead o c s wi h he his o ical p og ession o he p o essional-manage ial employee a e – whose geog aphy we conside a good indica o o social pola isa ion – b oken down a he local council le el (Figu e 20.6). The maps show inc easing a es o e he las h ee decades, con i ming he economic g ow h and specialisa ion o bo h 20 See he websi e o he managemen company o he de elopmen ad e ising: h p://sgal. al adelisboa.com/en/. Figu e 20.6 c s and p o essional-manage ial employee a e, 1991, 2001 and 2011 in Lisbon (abo e) and Cascais (below) Sou ce: ou elabo a ion; da a on p o essional-manage ial employees, Ins i u o Nacional de Es a ís ica, bg i censuses. 498 ambiguous inclusions: inside ou , ou side in ci ies. The pa e ns o pola isa ion, howe e , a e qui e di e en . In Cascais he e is s ong pola isa ion and, in ac , socio-economic seg ega ion, be ween he weal hie coas al a eas and he inne ones, wi h ga ed-communi y-like c s almos exclusi ely concen a ed in he leas weal hy a eas. Though one eason mus be he a ailabili y o mo e land o mul i- illa de elopmen s, he pe cep ion o hese a eas as he leas homogeneous and he p esence, o ins ance, o la ge social housing neighbou hoods, mus play a signi ican ole in he p oduc ion o mo e o i ied es a es. In Lisbon, pola isa ion pa e ns a e e y di e en , and show a complex blend o di e en si ua ions, wi h di e en axes o weal h, much mo e spa ially mixed wi h a eas less economically de eloped and social housing neighbou hoods. Though ea lie s ages o c de elopmen a e loca ed, as we sugges ed, in weal hie zones, he e has been a p og essi e sp ead in o di e en a eas. discussion: (un)planned spaces o inclusion/exclusion Ou mapping in Lisbon, Cascais and Ba ei o has con i med he exis ence o a numbe o peculia ai s o c s and hei p oduc ion in Po ugal. Wi h ega d o he ole o c s and simila de elopmen s in eshaping u ban and me opoli an e i o ies, while ga ed communi ies a e gene ally conside ed o be a subu ban phenomenon, we ha e seen how hey can become, and indeed a e mos ly used in me opoli an Lisbon as, ins umen s o es uc u ing dense u ban a eas. This is pa icula ly e iden in a cen al, dense ci y such Lisbon, bu e en in Cascais, in coexis ence wi h ypical examples o “p ime” subu ban de elopmen . Wi h ega d o empo al s ages, we ha e seen he c phenomenon booming a e 1995. Though we ha e no de ini i e da a o make conclusions on causal ela ionships, his aises he ques ion as o whe he he legisla i e and planning changes in ended and did egula ise, o a he p oduce, he c phenomenon in he i s place. Mi o ing discussions abou gen i ica ion (Hamne 1991), i is a he in e sec ion o supply and demand side explana ions ha his discussion should ocus. Fo ou pu poses, su ice o say ha , du ing he ea ly 1990s, na ional legisla ion and municipal planning (in Lisbon and Cascais) did open up he space o c s, a he e y leas . Besides emphasising he peculia i ies o he phenomenon in me opoli an Lisbon, we adop ed a compa a i e pe spec i e in o de o discuss o he explana o y dimensions. The di e ences among ou h ee cases a e inclusi e communi ies, exclusiona y ci y, planning n/a? 499 signi ican and add laye s o he discussion o he condi ions necessa y o c de elopmen s. Cascais is, o e all and despi e he abo emen ioned peculia i ies, he case whe e we ound a mo e ypical pa e n o ga ed communi ies, us s yle: he p esence o la ge, subu ban mul i- illa c s p o ided wi h ameni ies and spa ially secluded om a seg ega ed u ban ab ic – indeed, an example o a weal hy, i (and!) unequal, own. In Lisbon, c s a e ins ead a componen o u ban es uc u ing p ocesses; and concen a ion (in weal hy a eas), di usion and “mix” pa e ns (in he case o Al a de Lisboa abo e all) also coexis due o he dense socio-economic pola isa ion, almos in absence o seg ega ion. Agains his backg ound, i is wo h coming back o he issue o he social p oduc ion o c s. In Lisbon, in absence o he physical seclusion om he u ban ab ic and o he ypical spa ial ameni ies o ga ed communi ies, his is la gely limi ed o he p o ision o secu i y, which we can now conside in pe spec i e. On he one hand, secu i y de ices a e no exac ly “ ough” in some cases: a good example is he condominium DueDomani – loca ed in Chelas, a dominan ly social housing dis ic – ad e ised as a condomínio semi-p i ado (semi-p i a e condominium) and su ounded by a one me e all ence wi hou locks (Tulumello 2015, 496). On he o he , du ing ou ieldwo k we ound a g ea numbe o condominiums wi h 24/7 pa olling and cc s, e y simila o neighbou ing c s bu no ad e ised as such – and he e o e no included on ou maps. I is e y ha d, when no impossible, o dis inguish be ween c s and non-c s in Lisbon, i no o he ac ha some a e labelled so. This makes us e lec on he ac ha c s in me opoli an Lisbon a e o en no hing mo e and no hing less han a p omo ional s a egy (supply side) and hence an ins umen o middle class sel - ep esen a ion (demand side). This should be kep in mind while hinking abou he absence o c s in Ba ei o. On he one hand, we canno downplay he ole o planning. As we discussed, he 1994 municipal mas e plan made he cons uc ion o mul i- dwelling c s impossible and, mo e gene ally, he dominance o a s a u o y app oach o land use managemen (Tulumello 2017c) may ha e had a ole in p e en ing semi-in o mal de elopmen s such c s. A he same ime, howe e , land use managemen ailed in p e en ing ano he dimension o in o mal u banisa ion, and speci ically illegal allo men s, which can be ound h oughou Ba ei o – o he ex en ha a special ask o ce o hei upg ading and egula isa ion exis s in he municipal planning depa men . Mo eo e , as 500 ambiguous inclusions: inside ou , ou side in he case o Lisbon shows, one does no eally need much mo e han ad e ising and some secu i y de ices o p oduce and sell c s. As such, o he dimensions mus come in o play, abo e all he ac ha Ba ei o is among he leas (possibly he leas ) sui able municipali y in he me opoli an egion o c s. As a as loca ion is conce ned, Ba ei o is among he municipali ies wo s se ed by highways and connec ions o Lisbon – being in he leas a ou able posi ion wi h ega d o he wo b idges connec ing he sou he n bank o he Ri e Tagus o cen al Lisbon. As a as socio-economic composi ion is conce ned, he his o y o Ba ei o, i s cha ac e isa ion as an indus ial hub and ecen pos -indus ial decline, ha e no helped in d awing he middle and uppe classes owa ds he municipali y. conclusion: om inclusion/exclusion o agmen a ion/pola isa ion Two dimensions s emming om he mapping o c s in me opoli an Lisbon a e pa icula ly use ul when conside ing he inclusion/exclusion dicho omy. Fi s , we ha e seen how some peculia ai s o he Po uguese planning adi ion and u banisa ion egula ions ha e impeded he c ea ion o p i a ely owned collec i e spaces in mode n imes, in he name o a mode nis concep ion o u ban planning and o land as public compe ence.21 We canno help bu no ice how his is in con as wi h he his o ical de elopmen o Po uguese ci ies h ough blocks (qua ei ões), whe e dwellings enclose open spaces ha can be sha ed o di ided among he single esidences (quin ais, backya ds). In a way, some c ypologies, and he legal amendmen s ha opened up he space o hem, ended up ( e)c ea ing he possibili y o a heal hy ins umen o u ban design and planning. As he planne in Lisbon eminded us, when public spaces a e al eady gene ous (as is he case in me opoli an Lisbon gene ally), p i a ely owned common open spaces a e good o bo h he esiden s (who enjoy he p o ec ed open spaces) and he ci y (in e ms o managemen o public space).22 21 The idealisa ion o he ci y as a con inuous, e dan public space pinpoin ed by high- ise dwellings was one o he cen al aspec s o Le Co busie ’s wo k. 22 See Tulumello (2015; 2017a, 88-91) o c i ical discussions o he implica ions o he mode nis concep ion o public space o mee ing o he ness and eelings o sa e y in he public space. inclusi e communi ies, exclusiona y ci y, planning n/a? 501 Second, we ha e seen how c s a e o en a simple “label” a ached o qui e o dina y eal es a e p oduc s. In hese cases, hen, he double mo emen o inclusion (among pee s) and exclusion (o o he s and om o he s) loses spa ial meaning, while social meaning becomes pu ely symbolic – and he e lies he s a kes peculia i y o he Po uguese case in he global a ena. On he one hand, his makes sense in me opoli an ab ics whe e spa ial seg ega ion is ela i ely low and he encoun e wi h o he ness is much mo e common han in hose con ex s whe e ga ed communi ies we e bo n and a e lou ishing. On he o he , we ind he e u he e idence o he seconda y, i ma ginal, ole o he cons uc ion o “communi y” in c p oduc ion, Po uguese s yle. Though hese a gumen s add some nuances o mains eam c i ical discussions abou ga ed communi ies as inclusi e/exclusiona y de ices, ou poin is no o unde play he c i ical implica ions o ga ed esiden ial de elopmen s o u ban jus ice. Fo ins ance, he ac ha he possibili y o collec i e p i a e spaces emains limi ed o hose who can a o d access o luxu y and semi-luxu y housing shows how c s, in sca cely seg ega ed con ex s, a e capable o eshaping inclusion/exclusion pa e ns o ce ain bene i s – such as access o p o ec ed open spaces – in complex and nonlinea ways. As such, a he han sea ching o “epis emologies o hope” (Pow 2015) wi h ega d o exclusiona y u ban de elopmen s, ou conce n is o deepen he unde s anding o he complex ways exclusion/inclusion is ( e)p oduced. Th ee agendas o empi ical esea ch and heo e ical/me hodological de elopmen eme ge. Fi s , we see li le engagemen wi h planning as an ins i u ional p ac ice in in e na ional li e a u e abou ga ed communi ies; in o he wo ds, wi h laws, no ms, and plans, p og ammes and policies. Ou wo k – and ha o Raposo (2002) –, howe e , has shown he cen ali y o na ional and local ins i u ional and egula o y amewo ks in he unde s anding o how c s can sp ead, and inclusion/exclusion pa e ns be eshaped, in di e en con ex s. Second, we ha e shown some limi s ha he inclusion/exclusion dicho omy i sel has in explaining socio-spa ial phenomena in con ex s whe e seg ega ion is low. In he me hodological app oach o semi-quan i a i e mapping, and in he heo e ical concep s o pola isa ion and agmen a ion, we he e o e ound he g ounds o a loose , bu no less igo ous, amewo k o esea ch. By isualising he mic o- ela ions o inequali y wi h ega d o u ban de elopmen (pola isa ion) and he c acks gene a ed by spa ial en i ies such as c s ( agmen a ion), we we e able o explo e he g ain o socio-spa ial 502 ambiguous inclusions: inside ou , ou side in phenomena on a scale ha adi ional seg ega ion analysis ails o unde s and. This opens up i ually unlimi ed possibili ies o u he discussion, o ins ance, in he s udy o p oximi y/connec i i y ela ions be ween c s and social housing neighbou hoods, o he concen a ion/di usion o c s and public space and acili ies. In conclusion, le us e u n o he ole o c s in ecen egene a ion and gen i ica ion p ocesses in Lisbon. As ou da a is comp ehensi e only up un il he pe iod 2013/2015, we a e ully awa e ha many changes may be aking place now ha Lisbon has suddenly become a cen e o a ac ion o ou is s, in e na ional s uden s, middle-class o eign pensione s and s a uppe s, bu also in e na ional eal es a e g oups and pla o m capi alism behemo hs such Ai Bnb (see Viei a e al. 2016). Amid a u bulen socio-economic es uc u ing ha s a ed om he his o ic neighbou hoods and is sp eading as owa ds and beyond he consolida ed u ban co e, new o ms o exclusiona y de elopmen a e eshaping he u ban ab ic: luxu y blocks o la s, p ime s uden apa men blocks,23 hyb id o ms o sho - en al dwellings, o gi e jus a ew examples. These new de elopmen s, hough inse ed in he u ban co e, a e c ea ing new o ms o inclusion/exclusion, which link he ci y o global ne wo ks o weal h, capi al and mobili y, while c ea ing local disconnec ions – e. g. by making housing una o dable o i ually e e y “ egula ” wo ke in Lisbon, o pushing o a social cleansing o his o ically mixed neighbou hoods. Agains his backd op, e en he scale o analysis ha we ha e p esen ed he e may become oo wide o explo e new dynamics o inclusion/exclusion – and he es less sea ch o new sys ema ic and ine-g ained me hodological app oaches is he e o e much needed. 23 Collegia e Ma ques de Pombal is a pe ec example, www.collegia e-ac.p /en/. inclusi e communi ies, exclusiona y ci y, planning n/a? 503 A ailable wo ks abou c s in he Lisbon me o, wi h he o al numbe o c s ound pe municipali y e ei a e al. 2001 aposo 2002 soa es 2003* aposo, co a 2007 (unpublished epo , da a published in aposo e al. 2012) pe ei a 2010 ulumello 2012 p esen s udy Time ame Be o e 2001 1985-1998 Be o e 2002 1985-2004 1985-2004 wi h ew upda es a e wa ds Be o e 2012 Be o e 2016 Type o da a Coun Coun and lis wi h u he da a App oxima e coun Coun and lis wi h u he da a Coun and lis wi h u he da a Coun , lis and map Coun and map Sou ce o da a O iginal esea ch O iginal esea ch Fe ei a e al. 2001, sligh ly upda ed Raposo 2002 and o iginal esea ch Raposo, Co a 2007, sligh ly upda ed Raposo 2002 and o iginal esea ch All he o me and o iginal esea ch Lisbon 62 23 45<x<90 38 ——— 49 81 Cascais 86 43 ~90 89 91 ——— 127 Amado a 1 1 1 1 ——— ——— ——— Lou es 1 0 1<x<22 1 ——— ——— ——— Ma a 10 5 1<x<22 18 ——— ——— ——— Odi elas 0 0 1 0 ——— ——— ——— Oei as 31 7 ~45 11 ——— ——— ——— Sin a 15 6 1<x<22 14 ——— ——— ——— V F anca de Xi a 1 0 1 2 ——— ——— ——— To al me opoli an no h 207 85 n/a 174 n/a n/a n/a Ba ei o 1 0 1 0 ——— ——— 0 Alcoche e 1 0 1<x<22 1 ——— ——— ——— Appendix 20.1 504 ambiguous inclusions: inside ou , ou side in e ei a e al. 2001 aposo 2002 soa es 2003* aposo, co a 2007 (unpublished epo , da a published in aposo e al. 2012) pe ei a 2010 ulumello 2012 p esen s udy Almada 7 1 1<x<22 5 ——— ——— ——— Mon ijo 2 0 1<x<22 0 ——— ——— ——— Palmela 6 3 1<x<22 4 ——— ——— ——— Seixal ——— 0 1<x<22 1 ——— ——— ——— Sesimb a 7 2 1<x<22 5 ——— ——— ——— Se úbal 10 6 1<x<22 8 ——— ——— ——— To al me opoli an sou h 34 12 n/a 24 n/a n/a n/a Me opoli an Lisbon 241 97 n/a 198 n/a n/a n/a * The CF numbe s a e ep esen ed on a map o he me opoli an egion wi h ci cles p opo ional o he coun o each municipali y, g ouped in classes, esul ing in he alues p esen ed he e. Con inua ion… Appendix inclusi e communi ies, exclusiona y ci y, planning n/a? 505 acknowledgemen s This chap e s ems om coope a ion be ween Simone Tulumello’s pos -doc p ojec (“Which “secu e ci ies”? A c i ical app oach o secu i y and eelings o ea in u ban planning in Sou he n Eu ope”, unded by c , s h/bpd/86394/2012) and he p ojec “expe s – Making sense o planning expe ise: housing policy and he ole o expe s in he P og ama Especial de Realojamen o (pe )” ( unded by c , p dc/a p-eu /4309/2014 c ). ____________________________________ simone ulumello Ins i u o de Ciências Sociais, Uni e sidade de Lisboa A . P o . Aníbal Be encou 9 — 1600-036 Lisboa, Po ugal simone. ulum[email p o ec ed] o cid.o g/0000-0002-6660-3432 ____________________________________ alessand o colombo isc e-iul, Dinâmia-ce colom[email p o ec ed]m o cid.o g/0000-0002-9910-8999 ____________________________________ e e ences agkün, A. A., T. Baykan. 2012. “Ga ed communi ies in Is anbul. The new walls o he ci y”. Town Planning Re iew, 83 (1): 87-109. amendola, G. 1997. La ci à pos mode na. Magie e pau e della me opoli con empo anea. Rome: La e za. a baci, S., J. Malhei os. 2010. “De-seg ega ion, pe iphe alisa ion and he social exclusion o immig an s: Sou he n Eu opean ci ies in he 1990s”. Jou nal o E hnic and Mig a ion S udies, 36 (2): 227-255. a kinson, R., J. Flin . 2004. “Fo ess uk? Ga ed communi ies, he spa ial e ol o he eli es and ime-space ajec o ies o seg ega ion”. Housing S udies, 19 (6): 875-892. bap is a, I. 2013. “The a els o c i iques o neolibe alism: U ban expe iences om he ‘bo de lands’”. U ban Geog aphy, 34 (5): 590-611. beau ega d, R. 2003 [1993]. Voices o Decline. The Pos wa Fa e o us Ci ies. New Yo k, Rou ledge. caldei a, T. 2000. Ci y o Walls. C ime, Seg ega ion, and Ci izenship in São Paulo. Be keley: Uni e si y o Cali o nia P ess. câncio, F. 2008 [2005]. Cidades sem Nome. C ónica da Condição Subu bana, 2nd ed. Lisbon: Tin a da China. de du en, N. L. 2006. “Planning à la ca e: The loca ion pa e ns o ga ed communi ies a ound Buenos Ai es in a decen alized planning con ex ”. In e na ional Jou nal o U ban and Regional Resea ch, 30 (2): 308-327.