When Care Faces Violence: Experience of Anticipatory Grief, Chronic Vigilance, Ambiguous Loss and Resentment Among Caregivers for Street Dogs—A Qualitative Study from Istanbul

1. Introduction

Across towns and cities in Turkey, the everyday survival of street dogs is sustained by a dispersed workforce of volunteers. Caregivers feed and monitor dogs, arrange ad hoc veterinary care, mediate neighborhood tensions, and respond to crises. Taken together, these routine practices function as an informal but indispensable care infrastructure. When it holds, minor injuries are addressed before they escalate, frictions are defused, and dogs are more likely to remain visible, tolerated, and alive. When it frays, avoidable suffering increases and everyday relations of coexistence deteriorate.

Recent legal and policy shifts that reframe street dogs primarily as a matter of public order have placed this infrastructure under acute strain. These changes do not operate only through formal procedures or episodic enforcement; they also reshape the conditions under which care can be practiced. Caregivers describe the present period as governed by chronic vigilance: continuously scanning municipal communications and social media, recalibrating routines, and anticipating the sudden disappearance of dogs they know. This vigilance is accompanied by anxiety, resentment, moral distress, and anticipatory grief. Crucially, these are not merely emotional “effects” of policy. They operate as practical forces that reorganize care—shaping whether caregivers intervene, how visible they allow themselves to be, how they relate to neighbors and authorities, and whether they mobilize help for injured or newly abandoned animals. In this sense, caregiver well-being is not external to animal welfare; it is one of its enabling conditions, shaping dogs’ access to food, first aid, continuity of monitoring, and everyday conflict mediation.

In the aftermath of these legal and political shifts, many participants describe declining expectations of municipal assistance. Complaint hotlines, shelter intake, and street-level responses are widely experienced as discretionary, slow, or opaque. At the same time, reliance on private veterinary clinics intensifies because they remain the most immediate and trusted point of assistance. Yet this reliance exposes a structural bottleneck. Many clinics—because of space, staffing, and biosecurity constraints—cannot hold dogs even briefly. The absence of short-term canine holding, often needed for only 24–48 hours to stabilize recovery or arrange transport and foster care, repeatedly interrupts care trajectories. As urgent cases stall for lack of a safe place to keep a dog, pressure is displaced back onto caregivers’ homes, finances, and emotional reserves, and conditions that might have been manageable more easily escalate into emergencies.

Although volunteer caregivers are central to sustaining humane coexistence, dominant debates on street dog governance continue to privilege formal systems—shelters, regulations, and complaint mechanisms—while treating community networks as peripheral. This framing obscures how much urban welfare is produced in the space between institutions: through informal monitoring, rapid response, and the everyday negotiation of neighborhood tolerance. The capacity of these networks, and their fragile connections to community veterinarians, shapes how many animals reach formal pathways, how many conflicts are resolved before becoming complaints, and how much strain is redistributed onto municipal services.

This article examines volunteer caregivers’ experiences in Istanbul with three aims. First, it documents how the post-amendment environment alters the texture of everyday care—sleep, attention, routines, visibility, and thresholds for intervention—without collecting or reporting operationally actionable detail. Second, it traces how affective strain translates into welfare consequences by following where care capacity contracts or holds, and by identifying the supports (and gaps) that shape these trajectories, including the role and limits of private veterinary clinics. Third, it frames these dynamics as infrastructural rather than individual, asking how municipal opacity, discretionary encounters, and missing “bridge” arrangements between street, clinic, and recovery redistribute risk and responsibility onto informal care networks.

The study adopts an interpretivist qualitative design grounded in 43 in-depth interviews and eight months of fieldwork. Analytically, it combines constructivist grounded theory and reflexive thematic analysis to develop an account of how governance becomes lived as an embodied and practical condition of care. The findings are organized into four interlocking themes: (1) Anticipatory Grief and Chronic Vigilance, in which realistic threat becomes a structuring condition that reorganizes attention, routine, and intervention thresholds; (2) Shifting Meanings of Care and Responsibility, as commitment is re-scaled into situational risk management rather than abandoned; (3) Encounters with Municipal Actors and Shelters, where discretionary, delay-prone, and opaque pathways make institutional contact itself a wager that caregivers must anticipate and manage; and (4) Veterinary Bottlenecks and Short-Term Holding Gaps, where care trajectories repeatedly stall because many clinics can treat dogs but cannot provide even brief holding, forcing caregivers to absorb recovery logistics into precarious domestic and informal spaces. Taken together, these themes show that caregiver well-being is constitutive of animal welfare in the city: it shapes continuity of monitoring, timeliness of response, and the everyday conflict mediation through which coexistence is sustained. By foregrounding volunteer caregiving as an informal yet indispensable urban care infrastructure—co-produced with veterinary actors and constrained by institutional uncertainty—the article contributes to interdisciplinary debates on urban animal welfare governance by specifying the mechanisms through which legal uncertainty becomes lived, and by identifying non-extractive institutional “bridges” that could reduce harm without increasing risk for caregivers or animals.

2. Materials and Methods

2.1. Research Design and Methodological Approach

This study employed an interpretivist qualitative design to examine how recent legal and policy shifts reshaped community caregiving for street dogs in Istanbul. The study was not designed to collect or report operational tactics, routinized schedules, precise locations, or evasion strategies. Instead, it focused on how care is experienced, interpreted, and reorganized under legal uncertainty and perceived threat.

The central analytic unit was the voluntary care situation, defined as the dynamic relationship among caregiver(s), particular dogs (or groups of dogs), and neighborhood context, as shaped by institutional responses (municipal services, shelters, complaint mechanisms) and by veterinary access. This unit of analysis enables attention to mechanisms linking affect to welfare: how emotions such as chronic vigilance, anxiety, resentment, moral distress, and anticipatory grief become translated into changes in care capacity and, in turn, into everyday outcomes for dogs.

Conceptually, the study treats community caregiving as an informal but indispensable urban care infrastructure and approaches caregiver burden as relational and infrastructural—shaped by institutional uncertainty, neighborhood dynamics, and bottlenecks in veterinary access—rather than as an individual disposition.

Two complementary qualitative approaches were used: constructivist grounded theory (CGT) and reflexive thematic analysis (RTA). CGT guided inductive concept development from participants’ terms and interpretive frameworks, supporting sensitivity to local moral vocabularies and street-level understandings of legality. RTA was then used to develop analytically coherent, practice-relevant themes connecting affect, relationships, neighborhood dynamics, and institutional constraints. Themes were treated as interpretive outputs generated through iterative engagement with the dataset rather than as static “buckets” of codes.

This methodological pairing supported an animal-centred analytic trajectory: the analysis moves from situated accounts (e.g., delayed intervention due to fear; stalled care due to lack of short-term holding; unpredictable hotline responses) to patterned configurations that matter for animal welfare—where and why care contracts or holds.

2.2. Participants, Sampling, Interviews

To prevent deductive disclosure in a polarized and surveilled field setting, I report research sites through deliberately coarse spatial identifiers rather than district names or micro-localities (Kaiser, 2009). Following established disclosure-risk guidance, place is treated as a potentially identifying attribute—especially in “small-world” networks where a combination of role, incident, and location can enable jigsaw identification—even when personal names are removed (Saunders et al., 2015). I therefore group interview contexts into four non-identifying urban fabrics—High-Density Inner-City Neighborhoods, Peri-Urban Neighborhoods, Non-residential Industrial & Infrastructure Sites, and Low-Density Coastal Neighborhoods—designed to preserve analytic variation in built form, infrastructural connectivity, and governance exposure while avoiding traceable place description. Consistent with Nespor’s (2000) analysis, coarsening place identifiers balances analytic value with ethical necessity, recognizing that place itself can be identifying (Nespor, 2000).

This typological approach also reflects urban theory that treats the city as produced through heterogeneous, extended processes of urbanization in which peripheral edges and infrastructure corridors are not “outside” urban life but central to how regulation, visibility, and institutional encounter are experienced (Brenner & Schmid, 2014, 2017). I treat non-residential, industrial and infrastructure sites as analytically distinct urban settings because they are organized primarily around circulation and controlled access—logistical imperatives that intensify security governance and discretionary enforcement, and reduce public legibility—conditions that restructure what care, response, and protection can practically become in everyday life (Larrinaga & Doucet, 2010).

Fieldwork was conducted between July 1, 2025 and November 1, 2025 across multiple urban settings in Istanbul. Given the polarized environment and deductive disclosure risk, field sites are reported through four coarse, morphology-based spatial identifiers rather than district names: high-density inner-city mixed-use neighborhoods (n=14), peri-urban expansion zones (n=13), industrial/infrastructure corridors (n=10), and low-density coastal/agricultural neighborhoods (n=6). These identifiers capture broad differences in built form, visibility, and typical governance interfaces while preserving participant anonymity(Wiles et al., 2008). This coarsening is an ethics-by-design strategy (Saunders et al., 2015) implemented to minimize deductive disclosure in a politically polarized context. In qualitative research, detailed locations, temporal markers, and routines can enable “jigsaw identification” even after names are removed (Kaiser, 2009; Nespor, 2000; Saunders et al., 2015; Wiles et al., 2008). Following anonymisation frameworks that treat place as a potential identifier (Nespor, 2000; Tilley & Woodthorpe, 2011), I omitted, blurred, or generalized specific sites and years of engagement, recording only approximate temporal ranges to protect participants and animals while maintaining analytic value(Saunders et al., 2015). For each participant, only approximate years of engagement were recorded (Imray Papineau, 2024; Loperena, 2016; Saunders et al., 2015). Spatial identifiers are intentionally coarse, analytic categories (e.g., inner-city mixed-use; peri-urban expansion; industrial/infrastructure corridors; low-density coastal/agricultural) and are not tied to districts, streets, or specific caregiving sites. “Years of involvement” are reported as approximate durations (and may be rounded) rather than exact dates, and no routinized locations, schedules, or other potentially traceable details are included in the Table 1.1

Interlocutors of this research are all adult community caregivers (≥18 years) who regularly engaged in non-remunerated care for street dogs. Care activities included feeding and monitoring, first aid and transport, ad hoc fostering, and informal mediation with neighbors and municipal actors. Inclusion criteria required: (i) regular engagement in street-dog care; (ii) ≥12 months of caregiving experience to support reflection across pre-/post-legal amendment conditions; and (iii) self-identification as an “animal lover,” a locally salient moral category used by participants to describe commitment to care. In order to maintain analytic focus on community caregivers and volunteers, institutional actors, municipal squads, animal shelter employees were excluded from sampling.

Sampling proceeded through purposive sampling to maximize heterogeneity by caregiving role, years of experience, and neighborhood type, followed by snowball sampling appropriate for socially networked populations without public sampling frames. To mitigate common snowball biases (e.g., network homophily, gatekeeper influence), recruitment was seeded through multiple unconnected networks and sample diversity was monitored iteratively. The final sample comprised 43 caregivers with varied roles and lengths of experience.

Initial contact occurred via public-facing advocacy emails and in-person introductions in public spaces. Recruitment scripts emphasized a welfare-first, do-no-harm orientation: interviews would avoid operational tactics, repeatable schedules, precise locations, or other actionable details, and such information would not be collected or published. Participants gave consent in oral or written form depending on their preference and perceived safety. Also, during the fieldwork, consent was treated as an ongoing process. At the start of each interview, participants’ rights to skip questions, to pause, reschedule the interview, or stop without explanation at any time were communicated in plain language. The anonymization plan was reiterated, including broad spatial/temporal identifiers, time-shifting, removal of micro-geographies, and use of composite vignettes. Limits of confidentiality were clarified, including how the researcher would respond if imminent risk to any animal is clearly articulated (Imray Papineau, 2024; Loperena, 2016; Saunders et al., 2015).

A total of forty-three semi-structured interviews were conducted in Turkish. Most interviews took place in neutral and quiet venues selected by participants. Unless explicitly requested, interviews were not conducted at or near routine caregiving sites to avoid intensifying strain, disrupting caregiving, or exposing habitual locations. Because recording heightened discomfort and was often declined, interviews were not audio-recorded. Data were documented through contemporaneous fieldnotes and expanded notes prepared shortly after each interview. Key summaries were read back at the end of interviews to confirm accuracy and participant comfort. Interviews were scheduled, where feasible, outside peak caregiving hours to reduce time pressure and decision fatigue.

The interview guide was designed to elicit forms of experience and ethical reasoning rather than operational detail. Questions were framed to avoid names, precise locations, dates, routes, or instructions; institutional encounters were discussed at the level of roles, procedures, and constraints rather than identifiable people or places.

One module examined everyday changes since the legal amendment, focusing on how daily routines and perceptions of safety shifted. Sample questions included: “Since the amendment, what has changed most in your day-to-day caregiving routine?” and “How has your sense of safety or willingness to intervene in public space changed, if at all?” A second module explored affective rhythms and bodily impacts, tracing chronic vigilance, anxiety, resentment, moral distress, and anticipatory grief as lived states. Sample questions included: “Can you describe what ‘being on alert’ feels like for you across an ordinary day or week?” and “Are there particular moments or triggers that intensify these feelings, and how do you cope when they peak?”

A third module addressed encounters that shaped outcomes for dogs, including complaint dynamics, street-level interactions, attempts to seek shelter intake, and constraints within veterinary care—without soliciting identifiable details. Sample questions included: “Can you walk me through a recent situation where an encounter with an institution or authority affected what was possible for a dog?” and “When you try to access veterinary care or other formal support, what kinds of constraints tend to shape the outcome?” A fourth module mapped care capacity and informal infrastructures, focusing on the social and material conditions that enable or constrain action (helpers, transport, credit arrangements, and neighborhood tolerance/hostility). Sample questions included: “What kinds of support networks or informal arrangements make caregiving possible right now?” and “What tends to be the first thing to break down—time, money, transport, cooperation—when pressure increases?”

A final module centered on relational bonds and the ethics of coexistence, inviting reflection on attachment, loss, and the moral reasoning that sustains care under threat. Sample questions included: “What does this dog—or this group of dogs—mean to you, and how has that relationship changed under current conditions?” and “When risks increase, how do you reason through what you owe to the dogs, to yourself, and to others around you?” Anchored but non-probing follow-up questions (e.g., “What made that day harder?”) were used to deepen reflection and the conversation. When participants began to provide operational specifics, the interview was redirected toward structural or experiential dimensions (e.g., “We don’t need the exact place or time—what was challenging about arranging care for that specific animal?”). Interviews ended with a brief de-escalation routine (neutral recap; invitation to revise, to add to, or to extract from what the interviewee had already shared), and general mental-health or legal support resources were shared when requested.

Observation was conducted as single-visit, low-risk scans in public settings—such as parks, open squares, and the edges of marketplaces—to contextualize interview accounts without engaging in route-tracing or routine-following. Observation was conducted as single-visit, low-risk scans in public settings—such as parks, open squares, and the edges of marketplaces—to contextualize interview accounts without engaging in route-tracing or routine-following. Fieldnotes focused on ambient signals available in public space (e.g., general public signage and clinic postings indicating species limits or boarding policies), interactional cues (including residents’ tone, distance, and forms of intervention toward dogs), and visible access bottlenecks (such as queues at clinics and posted requirements that can exclude unowned animals from treatment). Observation did not involve engagement with municipal staff, collection of municipal documentation, or recording of identifiable institutional interactions.

To prevent reconstruction of routines, fieldnotes did not include routes, vehicle plates, clinic names, or precise timestamps. If an emergent scene risked exposure (e.g., a recognizable incident circulating publicly), note-taking ceased. During heightened enforcement periods (e.g., visible shelter vehicles, increased complaint traffic, reports of collection sweeps), in-person observation was paused and remote follow-ups were prioritized.

Interviews also followed trauma- and stress-informed principles insofar as supporting participant emotional safety matters, and reducing stress and trauma in animal care work can shape interactional dynamics (Stevenson & Morales, 2022a). Visible signs of distress (primarily, but not limited to, voice strain, prolonged silence, rapid speech) prompted breaks, topic shifts, or early termination. Repeatable specifics were not recorded; highly identifying detail, if it entered notes, was paraphrased into non-traceable form or removed during same-day expansion. No participant-provided documents, screenshots, videos, chat logs, complaint records, or evidentiary materials were collected or retained. In addition, the study did not draw on municipal correspondence, official case files, hotline logs, or exchanges with municipal/shelter personnel; institutional dynamics were documented only through participants’ accounts and low-risk public observation at the level of roles, procedures, and constraints. To mitigate power asymmetries, interviews were conducted in participants’ preferred vernacular; prompts were short and open-ended; and decisions made under pressure were treated non-judgmentally. When asked for advice, problem-solving was deferred until the end and pursued only if explicitly invited. Reciprocity was offered without identification risk (e.g., generic summaries on caregiver stress and well-being or small, non-traceable support gestures not tied to specific individuals or locations), and no meetings were posted or tagged on social media.

Participants were assigned non-identifying codes at point of contact; no real names were used in analytic files. Digital materials were stored on encrypted drives, with consent records stored separately. Participants were assigned non-identifying codes at point of contact; no real names were used in analytic files. Digital materials were stored on encrypted drives, with consent records stored separately. Because the article uses direct quotations, excerpts were selected and edited through a disclosure-risk screen: quotations were kept brief; references to micro-sites, distinctive incidents, and identifiable third parties were removed; and temporal or contextual details were generalized where needed. When a passage was analytically important but carried traceability risk, it was either paraphrased or incorporated into a composite vignette. Fieldnotes were expanded within 24–48 hours to capture contextual detail and reflexive observations (e.g., fatigue, positionality, social desirability dynamics), and temporal references were time-shifted by several weeks prior to integration into the analytic corpus. Composite vignettes were used where analytic importance intersected with identifiability risk, merging elements across multiple accounts, sites, and times to convey recurring configurations without exposing individuals, animals, or locations.

Fieldnotes were expanded within 24–48 hours to capture contextual detail and reflexive observations (e.g., fatigue, positionality, social desirability dynamics), and temporal references were time-shifted by several weeks prior to integration into the analytic corpus. Composite vignettes were used where analytic importance intersected with traceability risk. Distinctive cases (e.g., recognizable conflicts; specific clinic policies) were represented through composites that merged elements across multiple accounts, sites, and times to convey recurring configurations without exposing individuals, animals, or locations.

2.3. Researcher Positionality and Reflexivity

I occupy a dual position as a scholar and a long-term animal rights advocate involved in rescue and caregiving work in Turkey. This location in the field afforded access, rapport, and interpretive depth, but it also created predictable interpretive risks, including confirmation bias, over-identification with caregivers’ accounts, and moments of role conflict—dilemmas that are well documented in research conducted in animal-advocacy settings where surveillance, stigma, and exposure risks shape what can be said and how it can be reported. My approach to reflexivity followed an “ethically important moments” orientation: rather than treating ethics as exhausted by procedural approval, I treated ethical tensions and positional effects as part of everyday research practice that required ongoing judgment and documentation (Upton, 2011).

In practical terms, I used reflexivity as an analytic method through memo-writing throughout fieldwork and analysis, bracketing my initial assumptions before coding sessions, and conducting peer debriefs with colleagues not involved in data collection. Where my advocacy proximity increased the risk that participants or locations could become recognizable, I treated “thinning” or generalizing detail as an explicit ethical design choice rather than an analytic omission, consistent with the well-established point that anonymisation often requires compromise—as Tilley (2011) argues, especially in dense networks and sensitive contexts (Guillemin & Gillam, 2004; Loperena, 2016; Tilley & Woodthorpe, 2011). These decisions were also guided by beyond-human and multispecies research ethics that foreground obligations not only to human participants, but also to animals and to the relational fields in which animal vulnerability is managed, contested, and politicized—domains that conventional institutional review may not fully capture (Imray Papineau, 2024; Oliver, 2021).

Finally, my positionality required attention to the affective and moral stakes of animal caregiving and advocacy work. Because animal protection, rescue and care contexts are associated with heightened occupational stress, secondary traumatic stress, and compassion fatigue, I treated the emotional tenor of interviews and the interpretive weight of distress narratives as part of the research conditions rather than as “noise,” while taking care not to convert empathetic alignment into analytic foreclosure (Stevenson & Morales, 2022a).

2.4. Data Analysis

Interview notes and observational fieldnotes were compiled into a single analytic corpus and analyzed iteratively using a combined approach of constructivist grounded theory (CGT) (Charmaz, 2017, 2013) and reflexive thematic analysis (RTA) (Braun & Clarke, 2019; Byrne, 2022). Analysis proceeded in two coding cycles. First-cycle descriptive and NVivo coding prioritized participants’ own idioms, affective vocabularies, and locally salient legal/policy terms, preserving emic formulations and the texture of lived experience. Second-cycle pattern coding then clustered these codes into higher-order analytic categories and candidate themes linking affective states, everyday governance encounters, relational context, and care capacity (e.g., recurrent descriptions of “checking the phone at 3 a.m.” coded as chronic vigilance; narrowed intervention as withdrawal; clinic species limits and boarding constraints as veterinary bottlenecks).

Following RTA principles, themes were treated as interpretive outputs developed through sustained engagement with the dataset rather than as fixed containers of codes. Theme development involved iterative drafting, revising, and refining through analytic memoing and constant comparison across cases. Negative cases were actively sought to qualify and sharpen interpretations (e.g., accounts where care capacity remained stable despite perceived threat; situations where neighborhood solidarity buffered distress). Where appropriate, interview accounts were situated using low-risk observation in public settings and triangulation across participants’ descriptions of institutional procedures and constraints, not to establish an external benchmark of “objectivity,” but to locate interpretations within the everyday landscapes of care and constraint. The study did not incorporate municipal correspondence, hotline records, shelter files, or exchanges with municipal or shelter personnel as data sources.

To balance resonance with safety, limited participant feedback was sought only at the level of theme summaries (excluding quotations and timelines). Qualitative data management and coding were conducted in NVivo 12 Plus. Credibility was supported through method–theory fit (constructivist grounded theory and reflexive thematic analysis), iterative memo trails, attention to negative cases, and triangulation across interviews and low-risk observation. Transferability is intentionally bounded by Istanbul’s urban context and Turkey’s legal environment. Finally, the deliberate under-reporting of operational detail is treated as an ethical constraint rather than a methodological weakness, consistent with the risk of deductive disclosure in small-world networks; thick description is therefore provided at the level of care infrastructures (community networks, veterinary bottlenecks, relational dynamics) rather than traceable routines.

2.5. Institutional Review Board Statement

This study was conducted in accordance with the Declaration of Helsinki, and approved by the Ethics Committee of Kadir Has University (protocol codes: E-82741295-600-92377, approved on 14 July 2024, and E-82741295-600-123260, approved on 15 July 2025).

Participants could pause or withdraw at any time and up to 7 days post-interview; oral consent was used when safer. To minimize risk, the study avoided collecting or retaining materials that could enable surveillance or punitive action (including municipal correspondence, complaint logs, or institutional case records) and relied on participant narratives and low-risk public observation reported in generalized form. Given the risk of deductive disclosure in dense networks, full transcripts and detailed fieldnotes cannot be made publicly available.

3. Results

Analysis of 43 in-depth interviews, supported by eight months of qualitative fieldwork, generated a set of interlocking themes that describe how caregiving for free-living dogs in Istanbul is reorganized under institutional uncertainty after the 2024 legal amendment. Rather than treating “emotion,” “policy,” and “infrastructure” as separate domains, participants narrated them as mutually constitutive: changes in governance were experienced through the body (sleep disruption, intrusive checking, hyperarousal), through altered routines (timing, visibility, intervention thresholds), and through constrained pathways of assistance (municipal delay, shelter opacity, veterinary bottlenecks). Across interviews, caregiver wellbeing emerged as a material condition of animal welfare—shaping the continuity of monitoring, the timeliness of response, and the ability to mediate conflict in everyday coexistence.

Themes were developed through iterative, reflexive coding that moved from descriptive and in vivo codes (e.g., “being on alert,” “timing it to the minute,” “checking all night,” “I had to drive past”) to pattern-level categories capturing recurrent processes (e.g., vigilance as a structuring condition; care as risk management; discretionary governance encounters; holding gaps that stall care trajectories). The resulting themes are presented below as analytically distinct but empirically entangled mechanisms that together explain how legal uncertainty becomes lived, how it reorganizes care practices, and how it produces patterned welfare consequences for dogs.

Themes Identified Through Reflexive Thematic Analysis

Theme 1: Affective Strain of Caregiving for Street Dogs: Anticipatory Grief, Chronic Vigilance, Ambiguous Loss

Participants consistently described the post-amendment period as governed by sustained hypervigilance: not a momentary alarm but an ongoing orientation to realistic threat, in which the possibility of sudden removal, disappearance, or loss became ambient in daily life. This resembles what traumatic-stress scholarship conceptualizes as continuous traumatic stress—danger that is not resolved and therefore cannot be narrated as safely “in the past”—and what anthropological accounts of chronic crisis describe as emergency becoming a durable temporal order. Vigilance was enacted through monitoring municipal communications, social media, and neighborhood rumor, yet participants emphasized that this was not simply information-gathering. It was experienced as an affective–somatic condition: disrupted sleep, intrusive checking, concentration difficulties, heightened startle responses, and persistent arousal—patterns that align with pathways described in research on compassion fatigue, secondary traumatic stress, and burnout among animal-care populations.

Crucially, vigilance reorganized care. Routine sustainment was repeatedly displaced by risk calibration: caregivers described weighing not only what a dog needed, but what action might trigger in a volatile environment. This reorientation shaped whether and how caregivers appeared in public, the timing and form of interventions, and decisions about whether to seek institutional help at all. Vigilance, in other words, functioned as an affective mechanism through which caregiving became reorganized into anticipatory risk assessment under chronic uncertainty.

Anticipatory grief was tightly coupled with this vigilance. Caregivers described loving familiar dogs while preparing—emotionally, practically, and ethically—for a loss that could arrive without warning and without closure. This “pre-loss” horizon resonates with empirical work on human–animal bonds showing that distress often intensifies prior to definitive separation, and that anticipatory grief and caregiver burden are conceptually distinguishable yet empirically intertwined strains that can shape monitoring practices and decision-making. It also overlaps with ambiguous loss (Boss et al., 2021; Harris, n.d.), insofar as disappearance sustains attachment without confirmation, and with disenfranchised grief scholarship, insofar as animal-related grief is frequently minimized or socially illegible—conditions that intensify and prolong suffering even as care continues. Participants’ excerpts show how micro-events (a dog missing for days, a sound or vehicle signaling possible removal, a weather alert) repeatedly reactivated this pre-loss posture, sustaining emergency as a chronic rhythm rather than an interruption.

Theme 2: Shifting Meanings of Care and Responsibility—From Commitment to Risk Management

Across interviews, “responsible care” was described as shifting from open-ended commitment—grounded in compassion, duty, and neighborhood obligation—toward a situational, calculative ethics shaped by the perceived risks of visibility, the threat of loss, and constrained support pathways. Importantly, this was not narrated as diminished concern for dogs. Rather, participants framed it as an enforced narrowing of obligation into what could be sustained under conditions where help-seeking might backfire and where being publicly recognizable as a caregiver could itself become costly.

This shift was repeatedly expressed through how caregivers described routine. Outside institutional settings, daily care was narrated as improvised coordination—often solitary—marked by continual hassle and the grind of “keeping things going.” Under chronic pressure, care became a negotiation of response-ability: the practical capacity to respond to animals’ needs while remaining within shifting limits of time, safety, reputational exposure, and institutional volatility. Participants’ language captured the moral cost of this recalibration: caring continued, but it was increasingly rationed and defensively organized. Encounters with unfamiliar or newly abandoned dogs became particularly fraught, with several participants describing guilt, nausea, or shame when they could not intervene—moments that accumulated into moral distress not as a single dilemma, but as a repeated experience of falling short of one’s own standards.

This theme also shows how responsibility was redistributed through informal economies and mediated forms of care. Some participants described shadow budgets, informal credit, covert resource rerouting, or creative income streams to sustain veterinary bills and feeding—strategies that kept care possible while also deepening exhaustion and moral accounting. Others described the ambivalence of making suffering visible to unlock resources, highlighting a persistent ethical tension when documentation becomes both a tool of survival and a source of discomfort. Overall, risk management became the form commitment took—not because commitment faded, but because conditions made expansive commitment increasingly unsustainable.

Theme 3: Encounters with Municipal Actors and Shelters—Discretion, Delay, and Opaque Pathways

Caregivers’ experiences of “the state” were most often mediated through ordinary interfaces: complaint hotlines, municipal field units, shelter intake desks, and informal backchannels. What made these encounters destabilizing was not only the stakes for dogs, but the perceived centrality of discretion—who responded, how cases were classified, whether intake was accepted, and what follow-up occurred. From the caregiver perspective, outcomes were often experienced as uneven and difficult to predict, producing an everyday sense that institutional pathways were opaque, delay-prone, and contingent.

This unpredictability became part of care practice. Participants described “calling” or seeking official help as a wager whose consequences could be supportive, negligible, or harmful. As a result, institutional contact itself became something to anticipate and manage, rather than a stable route to assistance. The practical effect was a narrowing of feasible intervention, particularly for high-urgency cases involving unfamiliar, newly abandoned, or injured dogs—precisely the cases where time-sensitive response matters most. In such accounts, opacity had a distinctive welfare signature: not only fewer successful referrals, but patterned slowing at critical moments when medical windows are short and when conflict escalation can still be prevented through timely mediation.

At the same time, caregivers described developing an “everyday cartography” of discretion—learning which routes felt less risky, which interactions were more likely to help, and where trust was possible. Yet this adaptive knowledge did not resolve the underlying problem: the same governance architecture that sometimes provides care also authorizes coercion, making “help” structurally hard to distinguish from threat in everyday life. These encounters therefore fed back into Theme 1 and Theme 2: intensifying vigilance, deepening moral distress, and pushing caregiving further toward cautious, selective, and survival-oriented practice.

Theme 4: Veterinary Bottlenecks and Short-Term Holding Gaps—Where Care Trajectories Stall

As municipal pathways were experienced as uncertain or slow, private veterinary clinics emerged as the most immediate and trusted point of intervention. This pivot was not a simple preference for “better care,” but a pragmatic adaptation to unpredictability: clinics were spaces where treatment could begin without procedural delay. Yet caregivers repeatedly emphasized that veterinary access was constrained not only by cost, but by logistics (transport, time, mobility restrictions) and by the absence of viable bridging infrastructure between street, clinic, and recovery.

The most consequential bottleneck was short-term holding. Many clinics could not keep dogs even briefly because of space, staffing, and biosecurity constraints, even when they could sometimes hold cats. As a result, care trajectories often stalled around a narrow temporal need—safe containment for 24–48 hours—while caregivers stabilized post-operative recovery, arranged transport, negotiated foster placement, or waited for municipal decisions that might not materialize. The barrier was frequently not clinical but infrastructural: a treatable case could unravel because there was nowhere for the dog to stay once the clinic visit ended.

Caregivers described compensating through precarious domestic and informal spaces at personal risk, alongside escalating financial strain and reliance on informal credit. These conditions expanded secondary labors—finding temporary space, coordinating transport, tracking recovery, maintaining documentation—while simultaneously intensifying moral distress, because caregivers were forced into scarcity decisions with direct welfare consequences. Over time, these repeated blockages contributed to narrowing and withdrawal: limiting routes, reducing proximity, avoiding witnessing need that could not be met, and—in some cases—curtailing attachment as self-protection. Here the welfare consequences were direct: delays in treatment, disrupted recovery, heightened risk of deterioration, and increased likelihood that manageable conditions became emergencies.

3.1. Anticipatory Grief, Chronic Vigilance and Living in Ambiguous Loss

Following the 2024 amendment to Turkey’s Animal Protection Law ((Official Gazette of the Republic of Turkey, 2004, 2024), caregivers described the post-amendment period as a condition of sustained hypervigilance: an orientation to realistic, ongoing threat in which the possibility of sudden municipal removal, disappearance, or loss became an ambient feature of everyday life rather than an episodic event. Analytically, this resembles what traumatic-stress scholars conceptualize as continuous traumatic stress—danger that is neither resolved nor safely “in the past”—and what anthropological work on chronic crisis frames as the normalization of emergency as a lived temporal order(Eagle & Kaminer, 2013). Their vigilance was enacted through continuous monitoring of municipal communications, social media posts, and neighborhood rumor; however, participants emphasized that this was not merely informational labor but an affective–somatic state (sleep disruption, intrusive checking, concentration difficulties, heightened startle responses) consistent with secondary traumatic stress and burnout pathways identified among people who care for animals (Monaghan et al., 2024). Research on compassion fatigue in animal-care populations—including volunteers—similarly shows that sustained job/care demands predict secondary traumatic stress and burnout, and that organizational strain in animal-facing work can manifest as persistent arousal, withdrawal, and reduced care capacity (Monaghan et al., 2024) (Figley & Roop, 2006; Hill et al., 2020; Monaghan et al., 2024) In this sense, participants’ accounts suggest that caregiver wellbeing is not a secondary outcome but a constitutive condition of animal welfare under institutional uncertainty, aligning with trauma-informed perspectives developed within animal protection and welfare sectors (Hill et al., 2020b)(Figley & Roop, 2006). Because loss was continually anticipated rather than conclusively realized, caregivers described distress that accumulated in the interval of “living in pre-loss”—a pattern that closely mirrors empirical work on human–animal bonds showing that anticipatory grief and caregiver burden often intensify before death or definitive separation, and operate as related but distinguishable dimensions of strain (Spitznagel et al., 2019, 2021a). This pre-loss distress also aligns with findings that owners’ anticipatory suffering can shape care practices and decision-making trajectories in serious-illness contexts, while simultaneously producing relational and emotional load that may “spill over” into veterinary encounters (Uccheddu, 2021). Interviews also resonate with empirical work on anticipatory grief and caregiver burden in human–animal bonds, in which distress is produced in the interval of “living in pre-loss” (Spitznagel et al., 2019, 2021b; Spitznagel & Carlson, 2019).

Chronic vigilance functioned not simply as an internal affective state but as a structuring condition that reorganized the situated practice of caring for street dogs in Istanbul. In contexts of realistic, ongoing threat, vigilance tends to operate as a form of continuous traumatic stress in which safety work is never completed, and attention is persistently tethered to the next potential escalation rather than to a bounded “after” of the event (Eagle & Kaminer, 2013; Stevenson & Morales, 2022a). Caregivers’ accounts suggest that this condition materially reconfigured care: routine sustainment was displaced by a constant labor of risk calibration that shaped when and how caregivers could appear in public, how they assessed the costs of seeking institutional support, and how they managed the emotional and social liabilities attached to being read as “out of place” or transgressive(Hill et al., 2020b; Vigh, 2008). This is consistent with adjacent animal welfare and shelter literatures showing that animal-care labor—especially under institutional strain—frequently produces embodied stress responses, hyperarousal, and cognitive load (e.g., disrupted sleep, intrusive checking, concentration difficulties), and that these burdens intensify where care is entangled with moral conflict, secondary traumatic stress, and weak organizational protections (Monaghan et al., 2024). In this sense, vigilance should be read not as an idiosyncratic psychology but as an affective mechanism through which caregiving is reorganized into continuous, anticipatory risk assessment within a chronic crisis environment (Hill et al., 2020b; Monaghan et al., 2024).

This chronic vigilance, moral distress, and burden is tightly coupled with a shifting everyday experience of governance among caregivers. In this context, governance is not an abstract policy backdrop but something encountered and interpreted through routine contact with municipal procedures, neighborhood surveillance, complaint systems, shelter intake practices, and the discretionary power of street-level officials—particularly after the 2024 amendment that reframed street dogs as a matter of public order. Because enforcement is experienced as opaque and uneven, uncertainty becomes an immediate, practical condition of care rather than a distant legal question. Caregivers therefore continually evaluate not only what a dog needs, but what an intervention might trigger—transforming ordinary acts such as feeding, monitoring, or facilitating first aid into moments of risk calibration. Decisions about whether to intervene, how visible to be when seeking veterinary support, or whether to remain involved at all are shaped by this procedural and discretionary environment. Hypervigilance and anticipatory grief emerge here less as generalized anxiety than as a situated response to the felt possibility that a dog’s disappearance can follow from an unpredictable institutional encounter rather than from illness or injury alone.

Anticipatory grief was tightly coupled with this vigilant orientation and was articulated through a temporal horizon of pre-loss: caregivers described loving familiar dogs while simultaneously preparing—emotionally, practically, and ethically—for disappearance, forced removal, or death that could arrive without warning. This dynamic closely mirrors empirical findings in human–animal bonds showing that distress is often generated before a definitive endpoint, and that anticipatory grief and caregiver burden—while related—are distinguishable dimensions of strain that can differently shape monitoring practices, decision-making, and day-to-day functioning (Currin-McCulloch et al., 2024; Kogan et al., 2024; Kogan & Cooney, 2023; Spain et al., 2019; Spitznagel & Carlson, 2023). It also overlaps with ambiguous loss (Boss et al., 2021), insofar as disappearance or unconfirmed removal sustains attachment without closure, and with disenfranchised grief(Corr, 1999) and animal-bereavement scholarship documenting how animal-related grief is frequently minimized or rendered socially illegible, which can intensify and prolong suffering even as care continues(Cleary et al., 2022; Pihkala & Aaltola, 2025).

What emerges here is not grief as an “after,” but as an ongoing mode of attachment under threat—a pre-loss affective posture that closely parallels how companion-animal guardians in veterinary serious-illness and end-of-life contexts describe distress unfolding alongside an anticipated loss, rather than only in its aftermath. In this literature, anticipatory grief and caregiver burden are consistently treated as conceptually distinct—yet empirically entangled—pressures that can escalate together and mutually intensify strain over time. In participants’ accounts, this anticipatory state was continuously reactivated through micro-events that signaled possible catastrophe:

“The dogs disappearing for a few days triggers a panic that ruins my ability to focus on anything else,” one caregiver explained (Participant #9).

While another described an embodied monitoring routine—“I check the perimeter fences twice a night, not just for intruders, but to ensure no dog got trapped or injured by the barbed wire.” (Participant #10)

Vigilance also became climatological and infrastructural: caregivers learned to read weather forecasts, temperature drops, storms, and flooding alerts as imminent risk signals, while simultaneously treating the city’s material systems—fences and barbed wire, drainage and shelter integrity, construction sites and permit activity, traffic patterns and municipal vehicles—as a shifting hazard landscape that required continual checking, repair, and pre-emptive adjustment to keep dogs alive, unseen, and in place.

“When the forecast shows extreme cold, I spend the entire night trying to locate old blankets, unable to sleep,” (Participant #9) and

“My phone is permanently set to the weather app’s severe storm warnings. I can’t sleep until I’ve physically verified every shelter is dry and intact.” (Participant #16)

These practices can be read as a form of everyday disaster anticipation: attention narrows onto early warning signs, the body remains in a state of heightened watchfulness, and ordinary time is reorganized around preventing a loss that is always possible but rarely resolvable—hallmarks of “continuous traumatic situations” and community hypervigilance under ongoing threat.(Nuttman-Shwartz & Shoval-Zuckerman, 2016). In such conditions, emergency becomes a chronic temporal order rather than an interruption, with vigilance sustained through disrupted sleep and persistent arousal that can further intensify distress and erode recovery(Vigh, 2008).

At the same time, caregivers’ anticipatory grief was not only about imminence but about the status of loss—often unresolved, socially unrecognized, and repetitively reopened—bringing it into conversation with ambiguous and disenfranchised grief frameworks. Where ambiguous loss intensifies grief through uncertainty and foreclosed closure, disenfranchised grief highlights how mourning becomes harder when it is minimized or rendered socially illegible, a dynamic widely documented in animal loss. Many participants described a paradoxical rhythm in which threat amplified attachment and urgency:

“Every morning… I search for keywords like ‘missing’… bracing myself for the worst” (Participant #31)—while sustaining care required restraint and constant calibration, from tightly disciplined timing (“I time my feeding down to the minute…” Participant #23) to hyper-attunement to danger (“I’ve memorized the sound of every car…” Participant #30). Over time, the accumulation of near-losses and confirmed losses appeared to produce exhaustion and affective depletion—consistent with animal-care and rescue research linking chronic exposure and high-stakes responsibility to secondary traumatic stress, burnout/compassion fatigue, and morally laden strain that can culminate in withdrawal and intentions to leave (Connolly & Cullen, 2018a; Hill et al., 2020b; Kemp et al., 2016; Monaghan et al., 2024; Otto et al., 2002).

By narrowing attention toward imminent threats and increasing caution around visibility and intervention, chronic vigilance plus anticipatory/ambiguous grieving reduced the consistency of monitoring and the timeliness of response; caregivers described how this could allow minor injuries to worsen before being addressed, delay the de-escalation of neighborhood tensions, and weaken the everyday continuity through which dogs remain reliably visible, supported, and safer within urban life—echoing animal-protection scholarship that treats worker/caregiver wellbeing as a material condition of care quality rather than a side issue (Stevenson & Morales, 2022a).

3.2. Shifting Meanings of Care and Responsibility: From Commitment to Risk Management

Participants’ accounts suggest that the meaning of “responsible care” shifted from an open-ended moral commitment—grounded in compassion, duty, and neighborhood obligation—toward a more calculative, situational and improvisational ethics shaped by constant exposure to violence, ambiguous loss and constraint facing the constant threat of street dogs being captured, confined and killed.

One way to understand this emergent experience of new ethics in the face of violence is to think through how it implicates the making of daily routine during the care of street dogs. Outside institutional settings like shelters, the affective reorganization of routine in street-dog care often takes the form of improvised—often solitary—daily coordination, a continual hassle shaped by the grind of “keeping things going.”(Jun, 2024). This kind of re-organization resonates with recent anthropological and organizational scholarship that treats animal protection not simply as a bundle of practices, but as a charged social field in which people are made into moral subjects—whose commitments are repeatedly recalibrated in relation to public legitimacy, institutional scripts, and the exhausting work of “doing the right thing” under surveillance. Ethnographies of animal welfare and rescue show how care is routinely pulled into wider moral and political economies: transnational welfare projects, for instance, hinge on “equivalence-making” moves that render human and animal well-being jointly legible and governable, while also re-sorting whose vulnerability counts and how responsibility is allocated. In the context of street animal shelters, Jun’s ethnography of a South Korean animal shelter traces how compassion becomes an organized labor regime—gendered, disciplined, and reputationally fragile—so that commitment is not only felt but continuously managed amid scrutiny, scandal, and the risk of delegitimation (Jun, 2024). Complementing this, institutional ethnographies of animal sheltering and protection foreground how frontline care is coordinated through procedural texts, metrics, and cross-agency mandates (e.g., “One Welfare”), often leaving workers to improvise ethically and practically inside institutional gaps and accountability pressures—redefining what “commitment” can realistically look like in practice when the rescue, protection and treatment of street dogs is at stake(Koralesky et al., 2022, 2023b).

In this framing, “care” appears less as a stable virtue than a situated achievement: an ongoing negotiation of response-ability—the practical capacity to respond to animals’ needs while staying within the shifting limits of time, safety, reputational exposure, and institutional volatility. That emphasis resonates with feminist care-ethics work on animals in organizations, which argues that responsibility is fundamentally relational but never “purely interpersonal”: what care can be, and who is able to carry it, is decisively conditioned by organizational arrangements, governance logics, and the distribution of risk and authority (Connolly & Cullen, 2018a). It also aligns with Haraway’s insistence that response-ability is forged in entanglement rather than guaranteed in advance—an ethic that demands attentiveness while recognizing that obligations are enacted under constraint, and therefore repeatedly tested by what situations make possible (Haraway, 2023). What caregivers described, then, was not waning concern but an enforced narrowing of obligation into ongoing risk management: care had to be re-scaled to what could be sustained when visibility itself could trigger sanction, when help-seeking might backfire, and when the “right” action was continuously exposed to external judgment.

An increasing body of ethnographic work on animal sheltering makes this dynamic explicit: Jun shows how interspecies compassion can be drawn into moralizing regimes of ‘pity’ that discipline care, expose it to co-optation, and render caregivers’ commitment simultaneously demanded and precarious—so that compassion becomes something to be organized, defended, and managed rather than simply expressed(Frommer & Arluke, 1999; Jun, 2024; Koralesky et al., 2022, 2023b; Tallberg & Jordan, 2022). Related organizational ethnography on “care-based animal dirty work” similarly demonstrates how animal care labor is shaped by tensions of powerlessness, secrecy, stigma, and legitimacy—conditions that reorganize everyday ethics around keeping the work doable and socially survivable (Tallberg & Jordan, 2022). In critical life-and-death situations, this enforced pragmatism becomes even more visible: Sturman and colleagues’ research on informal community animal-care during Australia’s “Black Summer” bushfires shows how, amid anthropocentric disaster systems and thin official support, responsibility is repeatedly rerouted through vigilance, contingency planning, improvisation, and constant assessment of what interventions are feasible without generating new harms (Sturman et al., 2025). Over time, such recalibration produces moral distress not as a single dilemma but as an accumulative experience of repeatedly falling short of one’s own standards—an outcome consistent with evidence that high emotional demands (including euthanasia exposure), limited resources, and weak organizational support structures erode wellbeing and increase turnover intentions in animal care work, thereby destabilizing the continuity of care itself (Lezon et al., 2025).

Care was increasingly described not as a steady moral disposition but as a practical negotiation of response-ability: the ability to respond within shifting limits of time, safety, visibility, and institutional volatility. Responsibility remained relational and affectively grounded, yet it was continually shaped by governance environments that redistribute risk and constrain what care can realistically accomplish (Connolly & Cullen, 2018b), 2018). Under these conditions, commitment did not diminish; it took the form of risk management. One participant captured this plainly: “I still care deeply, but the care I can give is now limited… I have to think… at every step… if my intervention will help the dog or just put me at risk, and that question is killing me” (Participant #44). Another described how care became saturated by consequence and calculation: “It used to be about dogs. Now, every single move, every minute you spend outside, everything has consequences. You have to calculate every action” (Participant #45). This shift echoes ethnographic accounts of sheltering and animal protection in which compassion is pulled into moralizing regimes and legitimacy contests, so that care becomes something to manage and defend rather than simply enact .(Jun, 2024; Koralesky et al., 2022, 2023; Salem & Rowan, 2003; Singh & Dave, 2015; Tallberg & Jordan, 2022).

That recalibration also changed the boundaries of obligation. Encounters with unfamiliar dogs in need became especially fraught. One caregiver described driving past a dog they did not know, feeling physically unwell afterward, and being “forced to limit” help to dogs they could “keep safe” (Participant #46). In practice, this re-sorted eligibility for intervention: care concentrated around dogs already embedded in known routines and fragile protection networks, while newly abandoned, injured, or less familiar dogs were more likely to be deferred, passed by, or shifted into indirect forms of support. The same kind of ethical narrowing appears in institutional ethnographies of shelter work, where triage is shaped not only by scarcity but by scrutiny, procedural demands, and the need to produce defensible decisions under contested norms (Arluke & Irvine, 2017; Koralesky et al., 2023b).

Daily life was reorganized around improvised solutions and constant trade-offs. Several accounts rewrote ordinary measures of success into care-capacity metrics: “My stall profits used to determine my week; now, how much food I can buy for the sickest dog is my only metric for success” (Participant #1). Another described street care as selection under scarcity: “My clinical training focused on precision, but street care is just risk management—deciding which dog gets the one chance” (Participant #4). Others emphasized covert redistribution—“the volume of protein scraps I can secretly divert” (Participant #8) or “client tips for food and medicine” (Participant #32)—suggesting that sustaining dogs increasingly depended on shadow budgets, informal credit, and quiet rerouting of resources. This kind of piecemeal funding and reputational vulnerability is consistent with anthropological work on rescue economies, where care is sustained through unstable finances and continuous moral accounting(Caldwell, 2025).

Care also moves through alliances—sometimes explicitly multispecies, sometimes mediated through public-making. One participant framed dogs as “unofficial colleagues,” initially tied to perimeter safety but later becoming deeply personal (Participant #20), showing how practical coexistence can thicken into attachment. Others converted creative labor into care infrastructures: “I use the sale of their portraits to pay their bills” (Participant #27), or “My goal became documenting their life, not saving them… by raising awareness” (Participant #34). At the same time, the ethics of mediated care appeared as a persistent tension: “I film their suffering in exchange for donations, and I constantly grapple with the ethics of that exchange” (Participant #42). These accounts sit close to anthropological analyses of animal protection as moral subject formation and legitimacy work, where caring often requires continual justification, public negotiation, and reputational maintenance(Jun, 2024; Stevenson & Morales, 2022b).

The emotional cost of chronic vigilance, moral strain and loss accumulates. Distress came less from a single difficult decision than from repeated moments of withholding, delaying, or narrowing help in ways that violated one’s own standards—while attachment remained intact (Participant #44; Participant #46). Research on animal-care labor similarly shows that high emotional demands under constrained support structures undermine wellbeing and contribute to burnout and withdrawal, with consequences for care continuity (Lezon et al., 2025). Under risk-saturated conditions, the most immediate welfare effect is contraction: the sphere of possible intervention shrinks, and the dogs most likely to fall outside are those without established relations and protections.

3.3. Encounters with Municipal Actors and Shelters: Discretion, Delay, and Opaque Pathways

Caregivers’ experiences of “the state” were most often routed through ordinary interfaces—hotlines, municipal field units, shelter intake desks, and informal backchannels—rather than through formal legal processes. What made these encounters so destabilizing was not only the material stakes for dogs, but the way outcomes appeared to hinge on discretion: who picked up, which unit arrived, how a case was classified, whether the shelter accepted intake, and what kind of follow-up (if any) was offered. The resulting sense of unpredictability aligns with classic accounts of street-level bureaucracy, where frontline actors effectively “make policy” through discretionary triage under conditions of constraint (Gilson, 2016).

In Turkey, many volunteer- and caregiving-centered qualitative studies likewise identify municipalities as a major bottleneck in street-animal governance, showing how street dog population control has been caught up in state-led security discourse (Özdoğan, 2024) and how volunteers frequently report uneven fulfillment of local responsibilities and inconsistent municipal action (Tandoğan, 2024). Building on this, Güçlü argues that the prevailing “population management” repertoire (roundups, shelters, and removal from shared spaces) operates as a politics of domestication that redraws the boundary between “home” and “street,” othering dogs who remain in commons as legitimate objects of containment and disposal (Güçlü, 2024). Within this framework, adoption is not a straightforward welfare solution but often a privatized transfer of care in which trauma and its afterlives are managed by adopters with little to no public support, while regulatory imaginaries around “proper” adoption risk reproducing patriarchal and heteronormative inequalities unless they are explicitly redesigned to transform them.

That discretionary variability became something caregivers had to anticipate and manage as part of everyday care. “Every single move, every minute I spend outside, everything has consequences. I feel like I have to calculate every step I take” (Participant #45). The “action” here includes calling: whether a report will bring timely assistance, be ignored, or trigger escalation. Another caregiver described the internal corrosion produced by this uncertainty: “I still care deeply, but the care I can give now is rationed; I have to think twice if my intervention will help the dog or just put me, or put the dog at risk, and that question is killing me” (Participant #44). Research on animal control/field services offers a useful parallel: even in contexts with different legal infrastructures, officers themselves describe the work as requiring discretion and moral negotiation (“doing what’s right” versus “what the law tells you to do”), which helps explain why citizen-facing outcomes can vary sharply across workers and situations (Moss et al., 2022).

One consequence was a narrowing of what counted as feasible intervention, especially when the institutional route was experienced as uncertain or slow. The sharpest moral strain emerged in encounters with unfamiliar, newly abandoned, or injured dogs—precisely the cases that tend to require rapid institutional response. “I saw a dog I didn’t know needing help, but I had to drive past. I felt sick. I am forced to limit my help to only the ones I know I can keep safe” (Participant #46). In practice, this re-sorted eligibility for intervention: care concentrated around dogs embedded in known routines and fragile protection networks, while “unknown” dogs were more likely to be deferred or shifted into indirect forms of support. The wider sheltering literature helps clarify how such narrowing is structurally produced: institutional ethnography shows intake and frontline shelter work being organized through standardized procedures, documentation requirements, and triage categories that can slow action and constrain what becomes “actionable” even when need is acute (Koralesky et al., 2022, 2023a, 2023c).

Because institutional pathways were experienced as delay-prone and difficult to read, care increasingly shifted into improvised infrastructures—shadow budgets, covert redistribution, and small-scale financial engineering designed to reduce reliance on municipal responsiveness. “My stall profits used to determine my week; now, how much food I can buy for the sickest dog is my only metric for success” Participant #1). “My clinical training focused on precision, but street care is just risk management—deciding which dog gets the one chance” (Participant #4). “My identity changed from fixing engines to fixing broken bones. One is paid, the other costs me everything” (Participant #6). Other accounts described covert resource routing as an everyday technique of keeping dogs alive: “the volume of protein scraps I can secretly divert” (Participant #8), “client tips for food and medicine” (Participant #32), “cheap vet credit over my own comfort” (Participant #13). These tactics sit squarely within what volunteer research in Turkey identifies as the lived outcome of municipal inadequacy: citizens end up carrying sterilization, feeding, treatment, and emergency response through informal arrangements because institutional support is uneven or unreliable (Özdoğan, 2024; Tandoğan, 2024; Yıldırım, 2024).

Encounters with municipal workers and shelters also produced shifting expectations and uneasy alliances. Caregivers described learning, over time, which individuals were responsive and which routes were safest—an everyday cartography of discretion rather than a transparent service pathway. Some volunteers reframe dogs within local ecologies of care and safety: “These dogs are our neighbors... they are also useful for the community.... Large dogs are making me feel safe, they keep the perimeter safe. Bond between me and them is so powerful. They keep me safe, I have to keep them safe (Participant #20). Others described a shift in priority away from municipal framings of “public order” toward an ethic of survival: “I was worried about aggression, but now my mission is less about public safety and purely about their survival” (Participant #10).

This distrust is reinforced by scholarship on urban animal governance showing that municipal interventions are routinely pulled between two mandates that do not sit comfortably together: “protection” (welfare, rescue, sheltering) and “control” (risk management, nuisance abatement, spatial removal). In Srinivasan’s analysis of dog law and practice, governance explicitly operates through this dual aim, so that care is administered through the same institutional architectures that authorize seizure, confinement, and disposal—making “help” structurally difficult to distinguish from coercion in everyday encounters (Srinivasan, 2013). Many recent research on street dogs similarly explores how street dogs become governable through logics of informality and colonial biopolitics that legitimize exclusion and “eviction,” producing a situation in which state attention often signals heightened vulnerability rather than control and safety (Narayanan, 2017)(Ashleigh Best, 2021; Deckha, 2015; Srinivasan, 2025), 2022). Ethnographic and institutional research on animal protection work helps to explain why this ambiguity is experienced by citizens as risk: frontline enforcement and sheltering decisions are organized by legal thresholds, documentation demands, and procedural constraints that can delay or block intervention, even when suffering of animals is undeniably explicit —so outcomes hinge on discretion as much as on need (Koralesky et al., 2022, 2023b). In Turkey-focused qualitative work with volunteers echoes these dynamics in a local register, describing local governments as inconsistent and often inadequate actors, with shelter inefficiencies and “nonlegal” practices feeding uncertainty about whether municipal involvement will alleviate harm or intensify it (Tandoğan, 2024).

As official routes of municipal squads became harder to predict, some caregivers redirected effort into mediated forms of care—documentation, public outreach, and fundraising—both to reduce dependence on discretionary gatekeepers and to redistribute responsibility outward: “My goal became documenting their life, not saving them, because I realized I could reach more people by raising awareness than by direct rescue” (Participant #34). Another participant stated, “My research led me to care for them. It’s an extension of preserving the history of this city’s relationship with its marginalized life.”(Participant #37).

Yet this shift was ethically ambivalent, especially where visibility became a condition for resources: “I spare part of my income every month to buy dog food, and to fund their care. As if my life is tuned to their pain and suffering” (Participant #40). The sharpest articulation of that ambivalence appears when suffering becomes the currency that unlocks help: “My care is inherently transactional. I film their suffering in exchange for donations, and I constantly grapple with the ethics of that exchange” (Participant #42).

Ethnographic and organizational studies of sheltering and animal protection make a crucial point: compassion is not “expressed” so much as processed—channeled through intake categories, adoption screening, euthanasia schedules, complaint-handling scripts, donor expectations, and reputational management. In organizational research on animal shelter work, staff describe emotionally difficult tasks (especially intake and euthanasia) as requiring continuous sensemaking—ongoing work to keep actions intelligible as caring even when they are painful, publicly contested, or procedurally constrained (Schabram & Maitlis, 2017). Across classic shelter ethnography, the legitimacy problem is often handled through moral accounting: surrenderers and workers displace guilt by shifting responsibility, producing narratives that stabilize the shelter’s moral order while also sharpening judgments about “good” and “bad” forms of care (Taylor, 2004).

That legitimacy contest is not only external (the “public gaze”) but also internal, structured by unequal voice and competing moral projects. Volunteer-centered ethnography shows how volunteers mobilize moral, relational, and reputational resources to challenge routinized killing and to reframe “shelter death” as avoidable—turning the shelter into a field of ongoing dispute about what responsible care should be (Frommer & Arluke, 1999). Institutional ethnography adds a governance layer by tracing how frontline sheltering is coordinated through texts and standardized pathways (call-centre protocols, documentation requirements, decision rules), which can make care feel simultaneously urgent and administratively delayed—less a spontaneous virtue than an action that must pass through gatekeeping infrastructures to become “legible” and authorized (Koralesky et al., 2022, 2023b; Neumann, 2010). And when you widen the lens beyond shelters to professionalized animal protection more broadly, ethnography shows how organizational roles (investigator, lobbyist, expert, “moderate” advocate) actively produce particular kinds of “moral subjects”—people trained to anticipate scrutiny, defend choices, and keep the organization’s credibility intact even when that credibility is purchased through compromise (Reed, 2025).

Institutional opacity of animal shelters also produced distancing tendencies and affective intensifies of refrain, withdrawal and isolation on the part of caregivers. “My contribution is purely financial leverage; I cannot risk physical interaction. This distance is both necessary and morally isolating” (Participant #30). Responsibility, in these accounts, becomes a managed portfolio of risk and cost: “Caring for the strays is a rebellion against the strict rules of my student life. It defines my moral stance” (Participant #15) one participant states, while another mentions the inevitability of risk-taking in the face of financial constraints: “Every fish scrap I give them is a calculated risk against my ability to pay my rent…I have to live through this risk. I don’t know if my efforts make any change, but it’s always a trade-off” (Participant #18). These accounts refer to how caregivers are being pushed into quasi-managerial roles to tackle with the stigma of taking care of street dogs, improvising systems of care around institutional bottlenecks—articulates how volunteer work in taking care of street dogs also registers frustration, perceived municipal noncompliance with duties, and reliance on informal care infrastructures (E. Ashforth et al., 2007).

When municipal and shelter pathways are experienced as discretionary and opaque, urgent needs are pushed into channels where time is routinely lost—through unanswered calls, repeated referrals, shifting eligibility criteria, and the slow work of making suffering legible in institutional terms. Frontline systems often hinge on whether a case can be translated into recognizable categories of distress and into the specific interventions those categories authorize; when that translation fails or stalls, the result is delay rather than care.Discretion is not a minor administrative detail here: street-level bureaucracy research shows how discretionary triage becomes the practical mechanism through which public services are rationed under constraint, producing uneven outcomes that feel arbitrary from the caregivers’ side. For caregivers, that uncertainty becomes part of the moral calculus of whether to ask for help at all, and it narrows intervention to what can be sustained without triggering escalation or additional harm—“the care I can give now is unavoidably rationed” (Participant #34) and “my help is limited only to dogs embedded in known routines and protection networks” (Participant #66). Another participant sighs and asks, “nobody knows how to help, what to do to rescue dogs that are seriously injured, unfamiliar, newly abandoned” (Participant #15).

Dogs that are unfamiliar are the ones who were formerly captured and forcibly displaced by the municipal squads. As repeated in the interviews, those are the dogs that are often left out of familiar networks of care, established (albeit improvised) alliances, devoid of any stable caregiver, prior documentation or medical records. These are the dogs for whom delay of any kind of medical help is most consequential, because the window for effective treatment or safe de-escalation is often very short. Work on shelter systems also shows what happens when animals arrive late into already strained infrastructures: veterinary workload rises, length-of-stay pressures increase, and the capacity to respond quickly is further compromised—conditions that can deepen welfare risk even when intake occurs(Yan & Teng, 2023). In that sense, opacity carries a specific welfare signature: not only fewer successful referrals, but highly consequential, patterned slowing of institutional intervention and response in rescuing street dogs for whom time matters most (Koralesky et al., 2023b).

3.4. Veterinary Bottlenecks and Short-Term Holding Gaps: Where Voluntary Care Trajectories Stall

In Turkey, there is no community veterinary service for street dogs—especially at night—and no admission-free emergency walk-in clinics. Municipally administered street animal shelters are overpopulated, lack veterinary medical equipment and rehabilitation capacity (YİĞİT et al., 2019). When municipal response arrives late—or when the likely consequences of “calling for help” is felt unclear by the volunteer—care trajectories often shifted toward private veterinary clinics as the most immediate and trusted point of intervention. This turn was not a simple preference for “better care,” but a pragmatic adaptation to uncertainty: when caregivers could not predict whether municipal involvement would stabilize a case or intensify harm, they sought spaces where treatment could begin without procedural delay. That pivot sits squarely within broader research on access to veterinary care, which consistently shows that delays emerge not only from cost but from the absence of reliable pathways, limited service capacity, and the frictions of navigating systems under time pressure(Lem, 2019).

Yet volunteers’ heavy dependence on private veterinary service immediately produces its own bottleneck: short-term holding. Caregivers repeatedly emphasized that many clinics could sometimes take cats, but could not accept dogs for even brief stays due to space, staffing, and biosecurity limits. The welfare crisis often hinged on a narrow temporal need—safe holding for 24–48 hours—while caregivers stabilized a dog post-op, arranged transport, negotiated a foster space, or waited for a municipal decision that might never arrive. The gap was rarely clinical in the strict sense; it was infrastructural. A case could be treatable, even straightforward, and still collapse because there was nowhere for the dog to be safely kept once the clinic visit ended.

That “nowhere” is often experienced as a daily logistical trap. Emergency transport itself is uncertain and requires risk management: “Finding someone with a car willing to navigate these blocks for an emergency transport is impossible. We are entirely isolated” (Participant #3). For some, the only mobility available was improvised and fragile—“My bike is the only transport option for small injuries - and I can only carry cats. Not dogs. For anything serious or anything happening to dogs, I’m helpless” (Participant #11)—or punishingly time-consuming: “I can’t use public transport; buses do not take street dogs… And I don’t have a car… Which means I have to walk hours to the closest vet that offers a discount, which takes my entire day” (Participant #21).

For many caregivers, even routine feeding and monitoring carries physical hazard—“I have to cross three lanes of heavy traffic to get to the main feeding spot. While carrying boxes and bags of dog food, and sometimes bottles of water… I look homeless… And it is really dangerous to cross the streets when no car stops for me” (Participant #26). These constraints mirror what the access-to-care literature repeatedly identifies: logistics and transport of animals in need are not secondary problems; they structure whether care happens at all, and whether it happens in time, for caregivers that live in underserved communities (LaVallee et al., 2017; Lem, 2019).

When municipal pathways felt unreliable, and clinics could not bridge holding, caregivers described compensating through domestic and informal spaces—often at significant personal risk. “My house is a constant rotation of recovering dogs, and my landlord is threatening eviction. My sanctuary is my biggest risk” (Participant #19). Others built ad hoc infrastructures: “I created a makeshift infirmary in an old shed, but it’s constantly threatened by property owners who want the land back” (Participant #6). Material storage itself became precarious—“My storage is a broken-down freezer on the side of the street. It’s unhygienic and a fire hazard, but it’s all I have” (Participant #40)—and theft or sabotage forced concealment: “All my supplies are stolen or vandalized every two months. The only solution is to hide them, which makes access impossible in an emergency” (Participant #30). In this environment, “care” is repeatedly stopped not by lack of willingness, but by the absence of a stable bridge between street, clinic, and recovery.

The shift to private clinics also expands a cycle of personal debt and informal credit. Care became increasingly tethered to fragile financial workarounds: “The informal credit system with the butcher shop is maxed out. Without that, the feeding operation stops” ( (Participant #8). “The cost of a single vaccine is a week’s worth of my groceries. I can’t sustain this without external support” (Participant #16).

One participant admit in plain terms: “I have to rely on over the counter (God knows may be smuggled) veterinary supplies because I cannot afford the official prices or the import duties” (Participant #42). Some manage risk of being entangled in the webs of private veterinary service, debt and being obliged to unreliable medical treatments through invisibility: “I have to pay for everything in cash to avoid a paper trail that could expose me to my family and my job” (Participant #32). Even attempts to formalize protection for assets collapse under cost: “I tried to set up a legal trust, but the lawyer demanded a fee I cannot pay. All my assets are personal and therefore at risk” (Participant #44).

This pattern closely matches research showing that financial barriers are the most frequently reported constraint on veterinary access, and that those constraints reliably produce delayed or foregone care—especially when animals are unowned or responsibility is carried by under-resourced rescuers (Pasteur et al., 2024). It also aligns with ethnographic work on animal rescue economies, where caregiving becomes inseparable from unstable accounting, credibility, and the constant task of making resources appear from nowhere (Caldwell, 2025).

As cases stalled, caregivers described a widening set of secondary labors: searching for temporary space, coordinating transport, negotiating informal support, tracking recovery, and maintaining documentation under conditions that made “administrative competence” feel like another scarce resource. “The data collection system I use is a series of hastily scribbled notes on napkins. I know the data is crucial, but I can’t find time to digitize it” (Participant #24).

“My documentation is on a single, old laptop that is constantly crashing. If I lose that, years of my work are gone” (Participant #34). Meanwhile, institutional “solutions” dangle and freeze. One caregiver resents, “I was promised a space in the municipal shelter, but the paperwork is permanently stuck in a bureaucratic loop” (Participant #35).

These are the mechanics through which care trajectories stall: not one big failure, but accumulative friction across time-sensitive steps. “The holding gap” also intensify moral distress and vigilance, because it forces caregivers to carry the consequences of triage more directly. Retreat and avoidance are sometimes described as self-protection rather than indifference: “During feeding, there are some streets I intentionally refuse to go and see the dogs, because I know I cannot manage the hunger and the suffering of the packs there” (Participant #2).

Another caregiver mentions deliberate withdrawal from care and intentional self-restraint: “I limit my route now, deliberately avoiding the docks where I know the suffering is worse, to maintain my ability to function” (Participant #14). Some caregivers withdraw from direct intervention after repeated failures: “I used to try to catch them for neutering, but after failing three times and seeing their fear, I retreated to just feeding” (Participant #5).

Many others narrow down proximity with dogs: “I only feed them from a distance now; getting too close means realizing how much more I should be doing, which I can’t afford” (Participant #9), or ration their time to manage their emotional capacities to survive the pain and suffering of street dogs in need: “I only allow myself to spend 10 minutes a day with them; any more and the emotional toll is too great to continue the next day” (Participant #20). These accounts match a broader literature on compassion fatigue and secondary traumatic stress in animal care, where sustained exposure to suffering under constrained capacity predicts burnout, withdrawal, and coping through emotional numbing (Monaghan et al., 2024).

In the most acute cases, the holding gap turned caregivers into reluctant arbiters of scarcity. “The ethical burden of deciding which dog gets the last dose of medicine is… sometimes…nauseating me… I just hate myself when I calculate which dogs should get the medicine… I avoid that decision as long as possible” (Participant #25).

“My biggest struggle is… about…whether I am prolonging their suffering or genuinely helping them. It’s a constant moral tax” (Participant #31). “I’ve become hardened to the deaths. I have to; otherwise, I would cease to continue my life… To be able to survive, I am trying not to feel or think about their dying” (Participant #16). “I only focus on the ones I can save. I have created a psychological wall against the other 90%, and I have to live with that” (Participant #36). Some caregivers describe the deliberate curtailment of attachment—“I have stopped naming them. I can’t afford the attachment; the grief is too much” (Participant #43)—or a retreat after failed attempts to bring dogs into private homes: “I tried to adopt one but had to return it, and the shame of that failure has made me withdraw from all direct physical contact since” (Participant #39). This is where the clinic-as-solution becomes ethically double-edged: it enables treatment, but without a holding bridge it transfers ongoing responsibility—medical, logistical, emotional—back onto people already operating at the edge of capacity.

Institutional threat compounds these bottlenecks. Many caregivers describe environments where reporting or insisting could jeopardize fragile access: “I constantly see signs of neglect and abuse from other caretakers, but I can’t intervene because I’ll lose my own precarious access to the area” (Participant #18). “The police know who I am and what I do. I have to let some illegal acts of animal cruelty go unreported, just to maintain a fragile truce with the authorities” (Participant #041).

The bottleneck is not limited to what a clinic can medically do. It is produced across the wider conditions that govern whether care can be carried safely and continuously: delayed municipal response, unclear handover procedures, the absence of lawful temporary spaces, and the risks attached to visible intervention. Within this landscape, short-term holding is pivotal rather than incidental. Evidence from shelter systems shows that even one or two nights in foster care can reduce stress and improve outcomes for dogs, indicating how brief periods of decompression and individualized attention can alter welfare trajectories. In street-dog cases, the same 24–48 hours window often decides whether post-treatment recovery is secure, whether infection control is feasible, and whether a treatable condition deteriorates into irreversible harm. When municipal units arrive late and private clinics cannot keep dogs even briefly, this window becomes the point at which care most often breaks down—pushing cases into delay, worsening prognosis, and escalating costs borne privately by caregivers and bodily by dogs, in line with wider veterinary access research linking barriers to delayed treatment and increased severity.

4. Discussion: Withdrawal from Care, Bitterness, Resentment

The results show that care for street dogs operates as a fragile infrastructure rather than a stable moral capacity. Caregivers in Istanbul describe enduring attachment and responsibility, yet the conditions that shape action—legal uncertainty, public scrutiny, unpredictable municipal response, and missing “bridge” options between street, clinic, and recovery—continually compress what care can practically become. The result is a patterned retreat: not an abandonment of care, but a narrowing of where, when, and for whom care remains possible. “There are sections of the zone I refuse to enter now, because I know I cannot manage the suffering of the packs there” (Participant #2).

For many interlocutors, distance becomes a strategy for continuing at all: “I only feed dogs from a distance now; getting too close means realizing how much more I should be doing, which I can’t afford… I simply can’t handle when the state is killing dogs” (Participant #9). Time is rationed for the same reason—“I only allow myself to spend 10 minutes a day with them” (Participant #20)—and emotional numbing appears less as indifference than as a condition of endurance. Caregivers do not narrate these shifts—distancing, withdrawal, restraint—as neglect. They narrate them as morally injurious, lived through guilt and grief, and sustained by a psychological splitting that they neither celebrate nor fully control: “I only focus on the ones I can save… I have to live with that” (Participant #36).

This contraction reshapes the moral geography of who remains reachable. As commitment shifts from open-ended responsiveness to risk-managed obligation, care concentrates around familiar dogs embedded in established routines, while unfamiliar dogs—newly abandoned, injured, or socially peripheral—fall outside a reduced radius of intervention. The pain of that boundary is explicit: “I saw a dog I didn’t know needing help, but I had to drive past… I am forced to limit my help to only the ones I know I can keep safe” (Participant #36). What makes this boundary durable is not only scarcity, but the risk that visibility brings consequences. When institutional pathways are opaque and discretionary, “calling” becomes a wager, and care is recalibrated around what can be done without provoking escalation. Rationing is therefore experienced as internal corrosion rather than a neutral practical choice. The welfare implications follow: reduced coverage, weaker early mediation of conflict and injury, and heightened vulnerability for dogs who are less monitored and less socially buffered.

Care persists through informal infrastructures, but these supports remain thin and easily exhausted. A single driver, a sympathetic neighbor, informal veterinary credit, a clinic exception, a temporary corner of a home—these micro-arrangements often determine whether help is possible. Yet the same accounts show how quickly they fray under fear, fatigue, and escalating cost: “Finding someone with a car… is impossible. We are entirely isolated” (Participant #3). “My care network consists of a single retired neighbor and a teenager who can’t drive” (Participant #17). “My credit cards… are all maxed out” (Participant #8). “The cost of a single vaccine is more than a week’s worth of my groceries” (Participant #16). As municipal response arrives late and clinics cannot hold dogs for the short interval that often decides recovery, caregivers absorb the holding gap into private life at high personal risk—“my landlord is threatening eviction” (ID-019)—and are pushed toward triage decisions that register physically: “deciding which dog gets the last dose of medicine is physically nauseating” (Participant #25). In this sense, infrastructural gaps do not simply delay care; they convert care into debt, secrecy, and continual improvisation.

A more concerning toll appears when this repeated compression of care does not stop at guilt and grief, but hardens into bitterness—and, over time, into resentment. In the accounts, bitterness is not a fleeting irritation; it is an affective accumulation produced by ongoing exposure to preventable suffering under conditions where care is institutionally, medically, and socially undermined. Caregivers describe feeling pushed into impossible choices, blamed for outcomes they cannot control, and forced to carry the practical and moral burden of governance failures. Frustration follows when every attempt to “do the right thing” is met with blocked pathways, discretionary gatekeeping, or the sense that the state’s actions actively undo the work of care. Restraint then becomes a defensive posture: not only to reduce risk to dogs, but to reduce the psychic toll of repeatedly encountering harms one cannot interrupt. This sequence—bitterness, frustration, restraint, incapacity, guilt—often culminates in resentment, a sharper and more socially consequential form of moral injury because it reorganizes relationships, not just emotions.

Resentment matters analytically because it is one of the mechanisms through which violence thins the very solidarities that make care possible. Where care networks are already precarious, resentment can erode trust, weaken reciprocity, and intensify withdrawal. Participants’ accounts suggest that this resentment is not directed only “upward” toward municipalities or hostile publics; it can also become lateral—toward other caregivers who appear to “do less,” toward organizations perceived as selective, or toward acquaintances who disengage when risk increases. Under these conditions, solidarity becomes harder to sustain: coordination feels costly, asking for help feels humiliating or futile, and information-sharing can shrink as caregivers retreat into smaller, safer circles. In practical terms, resentment accelerates the privatization of responsibility: care becomes more individualized, more secretive, and more tightly bounded, even when caregivers continue to care deeply. The outcome is not simply emotional exhaustion; it is a degradation of collective capacity.

This is where the paper’s central claim sharpens: violence does not only harm dogs through direct removal, injury, or death; it also harms dogs indirectly by corroding the social ecology of care. When caregivers anticipate that visibility invites punishment, when clinical pathways cannot hold animals through short recovery windows, and when municipal action is inconsistent or perceived as hostile, the everyday work of care is forced into a posture of concealment and triage. Bitterness and resentment are not “secondary feelings” in this context; they become part of the breakdown of the care infrastructures. As Frommer et al. argues, caregivers’ frustration easily relocates onto other social networks, publics and other caregivers (Frommer & Arluke, 1999). As resentment spreads, networks thin, early interventions are delayed, and fewer people remain willing to take on cases beyond the small set of familiar dogs. Some caregivers are frustrated also due to unfair judgements, harsh criticisms and moral certainty of some caregivers, which in turn breaks down solidarities and possibilities of networking for collective caregiving ((Taylor, 2004). It is almost always the least visible dogs that pay the price of moral policing and breakdown of solidatiries. When there are too many conflicts around what is “good” or “bad” care, some dogs fall outside the radius of care—not because caregivers stop caring, but because the conditions of caring produce a social and emotional contraction that care cannot outwork.

What allows care to continue at all are small solidarities that momentarily stabilize collapsing chains: a neighbor’s help, a one-time transport, a clinic bending a rule, a small flow of food or money. These gestures do not solve the structural conditions described in Section 3.1, Section 3.2, Section 3.3 and Section 3.4, but they prevent some crises from cascading and they hold caregivers in the work for another day. At the same time, they remain precarious by design: when one person withdraws, one resource disappears, or one space closes, a whole care pathway collapses. The overall picture is therefore not only one of strain, but of a system that repeatedly privatizes responsibility while narrowing the practical space in which human-animal coexistence outside home can be sustained. Where micro-solidarities persist, they buffer distress and reduce immediate welfare harm; where they erode—through fear, fatigue, and especially resentment—delay expands, withdrawal deepens, and the collective capacity to keep dogs safe is progressively dismantled (Phillips & Gunter, 2024).

In this sense, withdrawal is not merely an individual coping strategy; it is a animal rights violations and welfare risk produced by institutional design. The findings therefore point to an urgent need for “bridge” care infrastructures that generate multiple challenges for care networks, and create secrecy, cycles of debt, and moral injury: predictable municipal responses, accessible night-time and emergency veterinary pathways, temporary holding and recovery options, and governance practices that do not criminalize or stigmatize caregiving in shared spaces. Without such supports, the most damaging effect of the current landscape may be the slow production of resentment—because resentment does not simply hurt caregivers; it fractures solidarity, and solidarity is one of the few remaining social infrastructures through which street dogs in Istanbul survive.

5. Conclusion

This article has treated volunteer caregiving for free-living street dogs in Istanbul not as a private moral inclination but as a public infrastructure: dispersed, informal, and routinely performing tasks that formal systems either cannot or will not do. The 2024 legal shift did not simply “change conditions” around this work. It reordered the field in which care becomes possible by amplifying threat, intensifying surveillance (formal and informal), and reclassifying everyday acts of attention—feeding, monitoring, transporting an injured dog—as potentially risky forms of visibility. In that context, chronic vigilance, anticipatory grief, and moral distress are not psychological “side effects.” They are governance effects, produced through institutional uncertainty and enforced discretion, and they shape welfare outcomes in direct, material ways.

Across four themes, the findings identify mechanisms through which violence toward street dogs is extended beyond overt capture, confinement, and killing. First, the post-amendment atmosphere makes loss feel both imminent and unconfirmable; disappearance becomes a routine possibility, and grief is lived in advance, repeatedly reopened. Second, responsibility is pushed from expansive commitment toward defensive risk management. Care does not disappear, but it is re-scaled to what feels survivable under threat. Third, municipal and shelter interfaces—experienced as discretionary, delay-prone, and opaque—turn institutional contact into a wager. Help-seeking is no longer reliably distinguishable from escalation. Fourth, reliance on private veterinary clinics, while often the fastest route to treatment, exposes a structural bottleneck: the absence of short-term holding. In practice, many care trajectories break not at the moment of diagnosis or treatment, but in the 24–48-hour interval where recovery needs safe containment, and where the lack of lawful bridging options forces responsibility back into private homes and precarious informal spaces.

The broader implication is political as much as empirical: caregiver well-being is an animal welfare issue. It is not an optional concern to be appended to “real” welfare metrics. When caregivers are compelled into permanent alertness, when grief becomes chronic and socially illegible, and when moral distress accumulates through repeated rationing and deferral, the result is predictable: thinner monitoring, delayed response, weaker early conflict mediation, and a narrowed radius of intervention. This contraction does not distribute evenly. It is the least protected dogs—unfamiliar, newly abandoned, injured, displaced—who fall first outside the shrinking sphere of care. What looks like an individual decision (“I had to drive past”) is better understood as an outcome of a hostile welfare regime that makes responsibility dangerous, time-consuming, and isolating.

The paper also traces a darker affective trajectory that matters for animal rights advocacy: the slow production of bitterness and resentment. These are not merely unfortunate feelings; they have infrastructural consequences. In Istanbul’s street-dog ecology, the capacity to keep animals alive depends on fragile solidarities—people sharing transport, credit, temporary space, information, and emotional endurance. When care is repeatedly undermined by institutional obstruction, when volunteers are blamed for outcomes they cannot control, and when visible caregiving is stigmatized or punished, resentment begins to corrode those solidarities. It can become lateral as well as vertical: directed not only at municipalities or hostile publics, but at other caregivers, organizations, and networks, especially under conditions of moral policing and reputational risk. In that sense, violence works indirectly by thinning collective capacity. It makes care more individualized, more secretive, more precarious—and the dogs pay the price.

From an animal rights perspective, the findings underscore a basic point that is often softened in welfare-oriented policy discourse: removal is not care, and killing cannot be justified as “management.” The governance architecture described by participants operates as a system of coercive displacement backed by discretionary power and institutional opacity. It produces not only immediate harms to dogs but also a longer-term dismantling of the social relations that sustain coexistence. That dismantling is politically useful to a removal-oriented regime: when networks fracture, fewer people can contest violence, fewer can witness it, fewer can mobilize timely assistance, and more suffering becomes normalized as “inevitable.”

What follows is not a generic call for “improvement” but a call for structural reversal. If municipalities and state institutions claim public responsibility for animals, that responsibility cannot be performed through punitive, opaque, and discretionary interfaces that offload risk onto volunteers while criminalizing or stigmatizing care in shared space. The most urgent “bridge” needs identified here are straightforward and welfare-critical: predictable, non-punitive municipal response channels; after-hours and emergency veterinary access for unowned animals; and short-term canine holding and recovery arrangements that do not force caregivers into illegal, unsafe, or eviction-risk domestic solutions. These are not technical add-ons. They are the difference between continuity of care and preventable collapse.

Methodologically, the study has been written under conditions where over-specificity can become harmful. To minimize deductive disclosure, it deliberately avoids operational detail and reports place through coarse categories. Interviews were not audio-recorded, and the analysis relies on expanded fieldnotes rather than transcripts. These constraints limit representational claims but are appropriate to the field’s surveillance and exposure risks; they are part of an ethics of not turning research into a resource for punitive governance.

Future research has to work inside a real ethical constraint: in a context where visibility can expose both people and animals to retaliation, “thick description” can become a liability. That does not mean, however, that institutional systems should remain analytically vague. The next step is to document—without producing operationally actionable detail—the bottlenecks and failure points that repeatedly surface in caregivers’ accounts: how discretionary municipal interfaces (hotlines, field units, shelter intake) generate delay and unpredictability; how handover procedures between street, clinic, shelter, and recovery are absent or unreliable; how veterinary capacity is unevenly distributed across time (especially nights and emergencies), space, and affordability; and how the lack of short-term canine holding functions as a predictable point of collapse for otherwise treatable cases. This kind of work can be done through careful institutional mapping and process analysis—tracking categories, timelines, eligibility rules, and resource constraints rather than routes, names, or micro-localities—and through triangulation with veterinary professionals and service providers in ways that protect identities. The aim is not to perfect a “better management” regime, but to make infrastructural responsibility legible: to identify where municipal services and veterinary systems systematically offload risk onto volunteers and onto dogs, and to specify what forms of bridging support would prevent preventable suffering without expanding surveillance or criminalizing care.