Safety is Not a Wall: Beyond Protection from Domestic Violence

Prasanna Pitigala Liyanage

Call her N.
The night she left, she packed her phone, her NIC, and a small bag.
A friend waited in a trishaw.
At the station, a statement; at court, a file; by morning, an order.
A line was drawn. A wall raised in law: no approach, no calls, no messages.

Walls can keep danger out.
They cannot cook rice, explain an absence, or put a child to sleep.
For a few days, she rested. She breathed. She planned.
Then the other walls came into view: rent, transport, school fees, a manager who wanted a reason without the story.
The protection order opened a corridor; it did not build rooms.

This is a design problem.

Her experience illustrates a documented gap: the PDVA provides a legal exit but often fails to address the material and social precarity that survivors face, a challenge noted in assessments of the PDVA’s implementation (Centre for Policy Alternatives 2020).

The tool interrupts harm for many. Yet when safety is reduced to distance, paperwork, and supervision, walls become the whole design, and care falls out of the drawing. The horizon is simple: freedom from violence, not a new architecture of surveillance.

This manifesto reads Sri Lanka’s Prevention of Domestic Violence Act (PDVA 2005) as designed space.
It treats legal rules as architecture, with plans, sections, and corridors, and asks how design can move law from surveillance to care.
It insists that designers, activists, judges, nurses, and bus drivers are co-architects of safety.

The Law: Corridors without rooms
Applications arrive in different ways.

Some are carried by a survivor.
Some are written by a police officer after a night of crisis.
Some are brought by a relative who has watched for years.

The court listens for a limited time. It aims to stabilise the situation until a fuller hearing can be held. At the full hearing, the court can maintain, vary, or discharge the order. The horizon remains the same: prevent further harm while ordinary life is rebuilt.

Enforcement depends on clarity and record-keeping. Clear language reduces conflict. Routine follow-up reduces fear. Because the device is spatial, it is legible to police, employers, and neighbours. That clarity is a strength. It is also a limit.

The statute creates a civil route in the Magistrates’ Courts.
It is designed for speed.
The court listens on limited time.
It aims to stabilise the situation until a full hearing can be held. If the court is satisfied with the evidence, it may issue an interim order. After a full hearing, it can issue a protection order with conditions.
The conditions are spatial and conduct-based: no direct or indirect contact, no approach within a set distance, no entry to designated places (home, workplace, shelter), no following or harassment, and no surveillance, threats, or contact through others, as directed by the court.
This is the wall.

Clarity helps police and neighbours know where the line is. But a line is not a home. The order opens a corridor where life might reorganise. The heavier lifts: rooms, services, cash flow, and cover lie beyond the court’s wall.

The interim protection order (IPO) is a temporary scaffold, the protection order is a longer wall, and the supplementary orders are doorways and windows that might be opened if someone asks. Yet the primary drawing is of distance, not of life. It is as if we designed a hospital with only corridors and no wards, no beds, no food court.

The PDVA sketches safety as absence: absence of contact, absence of approach, absence of harm. It rarely sketches safety as presence: presence of cash, shelter, childcare, community. That absence is not incidental. It is a design failure in the architecture of safety.

What the numbers say (and what they don’t)
Sri Lanka’s Women’s Wellbeing Survey 2019 (Department of Census and Statistics 2020) was the country’s first national study on violence against women, covering all 25 districts. It found that one in five ever-partnered women (20.4%) had experienced physical and/or sexual violence by an intimate partner. Nearly two in five (39.8%) had faced physical, sexual, emotional, or economic violence, or controlling behaviours, in their lifetime.

Economic abuse was widespread: 18.1% reported partners taking their earnings, refusing money for household expenses, or prohibiting them from work. The most common form of violence recorded was controlling behaviour (19.1%), a measure of how little agency women have in their daily decisions. The Centre for Policy Alternatives (2020) also critiques the law’s original formulation for its failure to name economic abuse as a standalone ground for relief explicitly, a significant oversight given that economic dependence is a primary barrier to leaving an abusive relationship.

The survey also showed how violence is normalised.

Almost half of women agreed that “a man should show he is the boss” (47.5%) or that “a good wife obeys her husband even if she disagrees” (46.5%).

Help-seeking patterns reveal the system’s design failures. Almost half (49.3%) of women who experienced sexual violence by a partner did not seek formal help, and one in five told no one at all. Among those who did, only 37.3% went to the police, and only 21.6% sought help from hospitals and health care centres. Over half (52.3%) said they did not leave because they did not want to leave their children.

Furthermore, the PDVA’s implementation is hampered by a system where survivors face immense social pressure to endure abuse and where institutional responses, including police mediation and judicial delays, can inadvertently prioritise familial preservation over survivor safety and autonomy (Centre for Policy Alternatives 2020).

The National Guideline for First Contact Point Health Care Providers (FHB 2019) recognises that health professionals may be the only point of contact for many survivors. It stresses that skilled providers can improve well-being, while lack of training can increase risk. It lists essential tasks, including listening, providing first-line support, urgent medical care, mental health care, and documenting evidence. It also acknowledges that most survivors attend health facilities for other reasons, which means the environment itself must encourage disclosure.

Numbers reveal prevalence, patterns, and silences. They explain exposure and under-reporting; they do not show how a survivor steps into a corridor and finds it narrowing or widening.
They cannot cook rice either.
But they tell us where the rooms are missing.

Critiques in practice
The problems are not only statistical; they are structural. The PDVA was drafted to move quickly, but speed without depth exposes survivors to fresh risks. Its weaknesses can be read as design failures.

1. The framing.
   The Act’s gender-neutral language was intended to reduce stigma, yet it obscures unequal power. In practice, many women are pushed toward remedies such as maintenance proceedings, which feel less stigmatising but provide weaker protection (Kodikara 2018). The Centre for Policy Alternatives (2020) also notes there is no comprehensive definition of “domestic violence”, and that the Act imports a Penal Code chapter for physical acts, which means acts such as marital rape are excluded from the definition. Emotional abuse must present as a “pattern” of a “serious nature”, with no clear guidance, and economic abuse is not named. These gaps invite narrow interpretation and leave major forms of control outside the frame. This is a failure of orientation: a corridor drawn without recognising who is most often forced to walk it.
2. The carceral horizon.
   The order interrupts harm but often stops there. It polices distance, but it does not scaffold survival. This mirrors what Goodmark (2018) calls the “carceral” approach: the state’s action begins and ends with an order, and the heavy work of life-building is left to the survivor, while economic and social supports are under-resourced. When the horizon is reduced to monitoring, the corridor becomes a trap, not a passage. This is a failure of imagination: safety reduced to absence, with no presence of support.
3. Latent design, rarely activated.
   Section 12 of the PDVA authorises the court to order counselling, small transfers, or escorted retrieval of belongings. On paper, the rooms are drawn. Too often, they stay unused. The powers exist but are not triggered. No bench prompt. No checklist that links a protection order to cash, housing, or childcare. Even when a door opens, the room is bare. No budget line. No housing partner. No escort staff. No childcare referral. This is a twin failure of use and supply. The blueprint holds rooms. They must be opened. They must be furnished.
4. The missing feedback loop.
   Protection orders are issued, but follow-up is inconsistent. Survivors are left to monitor their own safety, breaches are often met with mediation rather than enforcement, and most cases of domestic violence are dealt with through mediation by the police, rather than through the PDVA (Centre for Policy Alternatives 2020). Without review and refinement, the design stagnates. This is a failure of iteration: the plan is drawn once and left untouched.

Taken together, these critiques reveal a law that sketches safety as a corridor, not a home. It interrupts violence but does not sustain life. To correct these design failures, we need new tests and prototypes, as well as ways to furnish the corridor with exits, supports, and networks.

Design philosophy: From corridors to rooms
Architecture is not just about walls; it is about rooms, routes, and the quality of light.
Design, here, is not decoration. It is a method: listening, mapping, prototyping, testing, iterating. It translates survivors’ stories into spatial interventions and material support.
Design invites questions:

What happens in the first 72 hours?

Who shows up by Day 10?

Where are the doors, windows, and light?

Three imperatives guide this design approach:

• Practice spatial intelligence. Read law as architecture. Identify corridors, doors, and dead ends; insert rooms. Light matters. Privacy matters. Sequence matters.

• Pay material attention. Treat SIM cards, bus passes, letters, pouches, and cash envelopes as design objects with form, function, and meaning. A bus pass can say “I belong”; a SIM card can say “I am not cut off”; a pouch can say “This story is mine.”

• Commit to iterative practice. Co-design with survivors and frontline providers. Learn through immersion (observing courts, clinics, and homes), reflection (mapping flows and pinch points), translation (prototyping objects and spaces), and iteration (testing, failing, revising). A PDVA corridor is a prototype, not a finished building.

The design test: Three questions
From years of work with survivors and design teaching, a three-question test has been distilled. It is to be taped above every desk, bench, and triage room:

1. Exits. Does the person have a legitimate path out – open at night and on weekends, not requiring perfect fluency or flawless documents? Is the next step written down, with a date?

2. Material support. Did cash, housing, transport, and childcare arrive in time and without humiliation? A SIM card within hours, a cash transfer within days, a childcare slot within the first week.

3. Collective networks. Are the names and numbers of three people and two offices written down, reachable, and accountable? Is there a witness map – a simple sketch of who answers where, updated at each review?

If the answer is no, the design has failed. The remedy is not more walls; it is more rooms, more light, more doors.

This test aligns with the UN Essential Services Package for Women and Girls Subject to Violence guidance. The package calls for coordinated, multi-sector responses across health, police, justice, and social services, emphasising that collective provision can mitigate harm and break cycles of violence (UN Women et al. 2015). It notes that essential services collectively support women’s safety and well-being, contributing to long-term outcomes such as poverty reduction and achieving the Sustainable Development Goals. The National Guideline for First Contact Point Health Care Providers echoes this by outlining health-sector responsibilities while highlighting that most survivors suffer in silence and require supportive environments to speak (FHB 2019). The design test makes these principles tangible and measurable in local practice.

Prototypes: What could change tomorrow
Design begins with small pilots and honest ledgers. Choose one court, one clinic, and one community; test for 90 days; publish what changed and what failed.

Courtroom prototype
Reconfigure the courtroom layout.
Stop forcing survivors to face perpetrators.
Use gentle lighting rather than harsh overhead glare.
Install signage in Sinhala, Tamil, and English.
Designate a quiet waiting room separated from public corridors.
Create a one-page checklist attached to each order, including housing referral, cash trigger, school/employer letter, and next appointment details.
Schedule reviews at Day 7 and Day 30.
Train staff to ask about rooms and routines, not just breaches.
Sections 12 (1)(c–g) of the PDVA already permit counselling, monitoring, and monetary orders; a checklist simply activates these latent provisions.

Clinic prototype
Record the survivor’s story once.
Map each answer to a service: injuries to care, sexual assault to examination, children to social services, and emotions to counselling.
Use trilingual forms; maintain an interpreter roster.
Follow up at Day 7 and Day 30, not just once.
Follow the Family Health Bureau (FHB 2019) guidelines’ essential tasks – identification, first-line support, care of injuries, sexual assault care, mental health, and documentation.

School and employer prototype
Accept quiet letters; adjust attendance or shifts without interrogation.
Appoint a liaison who answers the phone and signs forms.
Protect the survivor’s privacy; do not demand details.
A school can treat a quiet letter as it would a medical certificate; an employer can treat it as a leave note.
This is part of the social‑services pillar in the Essential Services Package for Women and Girls Subject to Violence (UN Women et al. 2015).

The first thirty days
These notes draw on decades of design practice with survivors and frontline staff across the Western and Uva Provinces, from middle-income flats in Colombo to line rooms on Uva’s tea estates, where Malaiyaha communities live.

Design unfolds in time. Here is a blueprint for the first month:

• Days 1–3: A safe room; transport; a phone or SIM; a small transfer for food and medicine; a quiet letter template ready to adapt.

• Days 4–10: Childcare coverage; a school liaison; leave and shift flexibility at work; a clinic visit; a registrar appointment for document replacement.

• Days 11–20: A housing referral; a plan for safe routes; a bank appointment for a safe account; second review of cash sufficiency.

• Days 21–30: Review supports; update the witness map; cut paperwork to the minimum; plan for longer-term housing or co‑living.

This timeline responds to survivors’ lived realities and aligns with the FHB guidelines (2019) call for coordinated, early, multi-sector interventions. When survivors can see the next step and receive support without humiliation, they are more likely to walk the corridor.

Measuring what matters
We tend to count orders. Counting only orders is like counting corridors without rooms. We need to count rooms, days, and answers.

• Rooms prepared and used: How many safe rooms and shelters were available and occupied?

• Days to first cash transfer: How quickly did cash arrive?

• Days to housing referral: Did survivors secure housing within two weeks?

• Calls returned within 24 hours: Responsiveness matters.

• Witness map engagement: Are support names still answering at Day 30?

• Number of times a story was retold: Aim for once.

The Essential Services Package for Women and Girls Subject to Violence emphasises timeliness, survivor-centred communication, and coordination. (UN Women et al. 2015). If we treat these metrics like building codes, we can design accountability into the system.

Reform horizons
The PDVA’s blueprint is broader than its daily footprint. Section 12 already authorises urgent monetary assistance, housing costs, counselling, and shelter placement. Reforms should make these supports automatic, with presumptive orders unless there is a good reason not to.

Statutory changes could mandate cross-sector case conferences at Day 7 and Day 30. They could embed privacy by design into police and court data flows. They could codify duties for employers and schools to accept quiet letters.

Yet law alone is not enough. A feminist, decolonial design perspective warns against replicating carceral logic. The goal is not to entangle survivors further in surveillance, but to build exits, supports, and networks.

Toward a caring commons
Safety is not a wall; safety is a home.
The PDVA gave us walls. Design must give us rooms, doors, ramps, and light.
Justice is not the imposition of order; it is the capacity to remain with disorder without abandoning care.
Counting orders is not the same as counting lives rebuilt. Budgets are moral documents; they show what we value.

If we centre exits, supports, and networks; if we build prototypes with survivors, judges, nurses, drivers, and designers; if we measure rooms instead of walls, we can move from a carceral horizon to a caring commons. Only then can the corridor become a home.

A call to action
This is a manifesto. It is meant to be used.

If you are a judge, schedule reviews at Day 7 and Day 30 and activate the monetary and housing provisions already in the law.
If you are a clinic director, reconfigure your waiting room, train your staff, and adopt trilingual forms.
If you are a teacher or employer, accept a quiet letter without questions and provide flexible hours.
If you work at a bank, waive emergency fees and issue safe accounts.
If you drive a bus, learn to recognise the sunflower pass and stop without comment.
If you design policies or spaces, practice spatial intelligence, pay material attention and commit to iteration.
This manifesto asks you not just to read but to redesign. The corridor is drawn. The rooms are yet to be built.

Prasanna Pitigala Liyanage is a senior lecturer in the Department of Integrated Design at the University of Moratuwa and a Chartered Architect.

Image credit: Prasanna Pitigala Liyanage

References

Centre for Policy Alternatives. (2020). Legal Reform to Combat Sexual and Gender-Based Violence. Colombo: CPA. Available at https://www.cpalanka.org/wp-content/uploads/2020/11/Law-Reform-to-combat-SGBV-PART-1-General-Centre-for-Policy-Alternatives.pdf

Department of Census and Statistics Sri Lanka. (2020). Women’s Wellbeing Survey-2019. Colombo: Department of Census and Statistics, UNFPA, Government of Canada. Available at https://srilanka.unfpa.org/en/publications/womens-wellbeing-survey-2019

Family Health Bureau (FHB). (2019). National Guideline for First Contact Point Health Care Providers: Health Sector Response to Gender Based Violence. Colombo: Ministry of Health, Nutrition and Indigenous Medicine, Family Health Bureau, UNFPA, British High Commission Colombo. Available at https://srilanka.unfpa.org/sites/default/files/pub-pdf/National%20Guidelines%20Booklet.pdf

Goodmark, Leigh. (2018). Decriminalizing domestic violence: A balanced policy approach to intimate partner violence. University of California Press.

Kodikara, Chulani. (2018). “Battered Wives or Dependent Mothers? Negotiating Familial Ideology in Law”. Violence Against Women, 24 (8): 901–921.

UN Women, UNFPA, WHO, UNDP, and UNODC. (2015). Essential Services Package for Women and Girls Subject to Violence. Available at https://www.unwomen.org/en/digital-library/publications/2015/12/essential-services-package-for-women-and-girls-subject-to-violence

You May Also Like…