What Counts as Violence

A framework for seeing the forms of harm that homicide statistics cannot capture

By Janiel McEwan, Economist and Researcher

Opening

It is six in the morning in Montego Bay and a woman is walking to catch a route taxi. She has already made three decisions before the sun is fully up. She has chosen a route she considers safer than the shorter one. She has left her good phone at home. She has told her sister what time to expect her call. None of this appears in any government report. No officer records it, no survey captures it in the moment it happens, and if you asked her whether she had experienced violence that morning, she would almost certainly say no. Nothing happened to her. She simply organized her life around the possibility that something might.

A few miles away, a shopkeeper is pulling down his grille an hour before he would like to, because after dark the street empties out and he would rather close early than close alone. A father in Spanish Town is filling out a form at a government office for the third time because the first two were lost, and he cannot afford another day away from work to do it again. None of these people has been shot, robbed, or assaulted. And yet each of them has made a calculation that most people in low-crime societies never have to make: how much of ordinary life must be forfeited in exchange for safety.

This is where the real question begins. Not whether Jamaica is violent, which is too crude a question to be useful, but where violence actually lives in a society, in what forms, felt by whom, counted by which institutions, and missed by which of them.

Jamaica's murder count fell to 673 in 2025, the lowest annual total in more than three decades, according to the Jamaica Constabulary Force.[1] That is a genuine achievement and this essay will not pretend otherwise. But a country's relationship with violence cannot be reduced to the number of bodies its police force counts in a calendar year. If it could, the woman walking to the taxi stand would have no reason to plan her route around the hour.

Part I: Galtung's Great Insight

In 1969, a Norwegian sociologist named Johan Galtung published an essay in the first volume of a small academic journal he had founded himself, the Journal of Peace Research. The essay was called "Violence, Peace, and Peace Research," and it did something that most academic papers never manage: it changed the terms of an entire field before the ink had dried.[2]

Galtung began from a deceptively simple observation. Violence, he argued, is present whenever human beings are influenced in a way that causes their actual physical or mental realization to fall short of their potential realization.[3] Put plainly, violence is whatever prevents a person from becoming what they were otherwise capable of becoming. A man who dies of a treatable illness because there was no doctor within reach has experienced something that, under Galtung's definition, sits on the same continuum as a man who dies of a gunshot wound, even though no one pulled a trigger and no one intended harm.

From this foundation Galtung built his most enduring distinction. Direct violence is what most people mean when they use the word: an identifiable actor causing identifiable harm to an identifiable victim. A stabbing. A shooting. A beating. It has a subject and an object, a before and an after, a name that can go into a police report.

Structural violence has none of that clarity, and that is precisely what makes it dangerous. It is violence built into the structure of a society itself, showing up as unequal power and, consequently, unequal life chances.[3] Nobody decides, on any given morning, to deny a child in rural Jamaica the same educational trajectory available to a child in an affluent Kingston suburb. No single actor commits that harm. It simply accumulates, distributed across institutions, budgets, and generations, until the gap between what that child could have become and what she is likely to become is every bit as real as a wound, only invisible and slow.

Galtung paired this with a second distinction that peace researchers still argue over today: negative peace, meaning simply the absence of direct violence, war, or open conflict, against positive peace, meaning the presence of social justice, meaning the active dismantling of the structures that produce harm.[2] A country can achieve negative peace, in the sense that nobody is currently shooting anybody, while remaining very far from positive peace. This is the exact fork in the road where Jamaica now stands, and it is worth sitting with that discomfort rather than rushing past it.

Twenty-one years later, in 1990, Galtung returned to the subject with a paper called simply "Cultural Violence," published in the same journal.[4] He defined cultural violence as any aspect of a culture that can be used to legitimize violence in its direct or structural form.[5] Religion, ideology, language, art, science: any of these can be recruited to make direct or structural harm look normal, inevitable, or simply not worth mentioning. Cultural violence does not itself kill or deprive. What it does is far more corrosive. It makes killing and deprivation feel unremarkable, or even correct.

Galtung pictured the three forms as a triangle, with causal energy flowing in every direction. Cultural violence legitimizes structural violence. Structural violence produces the frustration and desperation that often erupts into direct violence. Direct violence, once normalized, hardens into new cultural narratives about who deserves protection and who does not.[5]

It is worth saying plainly what Galtung's framework is not. It is not a claim that every hardship is violence, and it is not a license to call any unwelcome social outcome an act of aggression. Even sympathetic readers of Galtung have noted that his definitions, taken too loosely, can become so broad that almost any social problem gets relabeled violence, at which point the word stops doing analytical work and starts doing rhetorical work instead. That is the trap this essay is trying to avoid, and it is the reason the framework proposed below insists on tests rather than assertions.

The honest question, then, is what Galtung's insight would look like if someone tried to operationalize it. Not just to name structural and cultural violence as categories, which Galtung already did with more rigor than most of his successors, but to build something a journalist, a policymaker, or an ordinary citizen could actually use to look at a real society and ask, systematically, where is the harm, who is bearing it, and can it be shown rather than merely felt.

That is the gap this essay tries to close.

Part II: The McEwan Index of Social Violence

What follows is an original analytical framework developed by The McEwan Index, built on Galtung's foundation but extended for a set of problems Galtung never had to confront: real-time data, competing official statistics, social media amplification, and a Caribbean state trying to measure its own progress against its own history. It is not an established academic index. It has not been peer-reviewed. It should be read as a proposal, a working tool offered for other researchers to test, criticize, and improve, in the same spirit that Galtung offered his own typology for others to extend.

The central question the framework tries to answer is this: how do we systematically identify violence in a society before we collapse the entire concept down to murder, assault, or war?

The framework organizes potential violence into nine dimensions. None of them is automatically violence. Each dimension identifies a zone where harm might be occurring, and each requires evidence before the label is applied.

Physical violence covers observable bodily harm: homicide, attempted murder, shootings, sexual violence, domestic violence, child abuse, police violence, deaths in custody, and armed conflict. This is the dimension every country already measures, and for good reason. Homicide is the most reliable violence statistic any government produces, because a body is difficult to hide and a death is difficult to reclassify. But reliability is not the same as completeness. A country can reduce its homicide count sharply while every other form of harm below remains untouched, and if homicide is the only dimension being watched, that country will believe itself far safer than it has actually become.

Structural violence, following Galtung, asks whether institutions and economic arrangements systematically prevent people from reaching their basic human potential: poverty, unemployment, unequal access to healthcare and education, inadequate housing, geographic and digital exclusion. Here the framework must be explicit about a line that gets crossed carelessly all the time. Poverty is not automatically violence. Inequality is not automatically violence. The question is not whether disadvantage exists, since disadvantage exists in every economy on earth, but whether that disadvantage is avoidable given the resources and knowledge a society actually possesses, and whether it is being reproduced by identifiable structures rather than by chance. A famine caused by drought in a society with no capacity to prevent it is a tragedy. A famine caused by drought in a society that exported food throughout the crisis is something closer to what Galtung meant.

Institutional violence is a category this framework treats as distinct from ordinary structural disadvantage, because it has a different mechanism. It asks what happens when institutions themselves, through action rather than mere absence, repeatedly produce avoidable harm: administrative exclusion, bureaucratic delay that costs people their entitlements, corruption, abuse of authority, excessive use of force, lack of accountability, unsafe public infrastructure. The test here cannot be "did the institution fail," because institutions fail constantly and most of that failure is ordinary inefficiency, not violence. The test has to be narrower: is the failure recurring, is it foreseeable, does it fall disproportionately on people with the least power to challenge it, and has the institution had the opportunity to correct it and declined to.

Economic violence asks when economic structures or practices impose severe and avoidable harm on the people subject to them: predatory lending, wage suppression, exploitative employment, exclusion from credit, forced migration driven by economic collapse. Again, ordinary market inequality is not violence. A worker earning less than a colleague with more experience has not been harmed in Galtung's sense. A worker trapped in a debt cycle by a lender who deliberately targeted her precisely because she had no alternative is a harder case, and closer to the line.

Psychological violence covers harm that may leave no visible injury: intimidation, coercive control, harassment, stalking, chronic fear, threats against journalists and witnesses, political intimidation. This is the dimension hardest to measure with any confidence, because self-reported fear is shaped by media exposure, personal history, and a dozen other factors that have nothing to do with actual risk. The honest answer is that psychological violence should be tracked through triangulated evidence, meaning victimization surveys, clinical and hospital data where available, and documented patterns of threat, rather than through any single number treated as definitive.

Cultural violence, again following Galtung directly, examines how language, religion, gender norms, class assumptions, and media narratives can normalize direct or structural harm. This dimension must be handled with particular care, because it is the one most easily abused to smear an entire group or belief system. The question is never "is this culture violent." The question is "what specific beliefs, in this specific context, make a specific form of harm look normal, deserved, or invisible."

Environmental violence considers whether unsafe water, air pollution, toxic exposure, and disaster vulnerability constitute a form of systemic harm, particularly when certain communities are made to bear a disproportionate share of environmental risk that they did not create and cannot easily escape. The framework uses this term descriptively rather than as a settled legal or moral category, since the scholarship on environmental justice is itself still working out how far the concept should extend.

Temporal violence is the most original and most speculative dimension in this framework, and it asks a genuinely difficult question: can harm be embedded in time itself. Childhood deprivation that produces adult disadvantage, intergenerational poverty, inherited trauma passed down without ever being named, decades of institutional neglect that compound into a single life outcome. The analytical challenge is that temporal violence is nearly impossible to separate cleanly from the immediate event that eventually expresses it. A young man's involvement in violence today may be the visible tip of thirty years of accumulated structural and psychological harm that no single agency ever measured as it happened. Treating that accumulation as a form of violence in its own right, distinct from whatever act finally makes it visible, is one of this framework's most contestable claims, and it is offered here as a hypothesis for other researchers to test rather than as settled fact.

Digital and informational violence covers the newest terrain: online harassment, doxxing, coordinated intimidation, algorithmic discrimination, and incitement amplified by social platforms. The evidence base here remains genuinely thin in a Caribbean context specifically, and this essay will not pretend otherwise. Where global research exists, it is cited. Where it does not, that gap is stated rather than filled with speculation.

Part III: The McEwan Violence Test

A framework this broad is useless without a discipline for deciding when a condition actually qualifies as violence rather than simply as hardship, unfairness, or bad luck. The McEwan Violence Test proposes ten criteria. A condition does not need to satisfy all ten to be treated as a form of violence, but the more of these that are present, and the more strongly, the stronger the claim.

1. Harm. Is there demonstrable physical, psychological, social, economic, or developmental harm, not merely discomfort or inconvenience.
2. Constraint. Does the condition materially restrict a person's ability to live, participate, develop, or exercise meaningful choice.
3. Pattern. Is the harm isolated, or does it recur in a identifiable pattern across time or population.
4. Power. Is there an identifiable imbalance of power, resources, authority, or opportunity between those who bear the harm and those positioned to prevent it.
5. Avoidability. Could the harm reasonably be prevented or substantially reduced given the knowledge and resources actually available, not some idealized standard no country meets.
6. Inequality. Does the harm fall disproportionately on particular populations rather than being distributed roughly evenly across society.
7. Legitimation. Are cultural, political, institutional, or media narratives being used to normalize, justify, or render the harm invisible.
8. Duration. Is the harm temporary, chronic, or intergenerational.
9. Institutionalization. Is the harm embedded in rules, incentives, or standard practice, rather than arising from a single actor's isolated decision.
10. Measurability. Can the harm be observed through credible empirical indicators, rather than resting on assertion alone.

These criteria are not weighted equally in this framework, and a weighted model that pretended otherwise would be manufacturing false precision. Harm and avoidability function as gatekeepers: without demonstrable, avoidable harm, a condition should not be called violence no matter how many of the other eight boxes it checks. The remaining criteria function as amplifiers, strengthening or weakening a claim that has already cleared the gate. This ordering is itself a judgment call, offered for debate rather than presented as proven.

The McEwan Violence Matrix

The nine dimensions and ten-point test can be organized along three axes that, taken together, make it possible to see how a single society can contain multiple forms of violence operating simultaneously without any of them showing up in the same government report.

The point of the matrix is not the diagram. It is the discipline of asking, for any society, which cells are being filled with data and which cells are being filled with silence. A country that measures the top-left cell obsessively while leaving the bottom-right cell entirely unmeasured will always believe it is safer than a fuller accounting would show.

Part IV: Jamaica's Numbers, Read Carefully

Jamaica's homicide trajectory over the past two decades is genuinely one of the more dramatic security stories in the hemisphere, and it deserves to be stated with precision rather than either triumphalism or dismissal.

The country recorded 1,680 murders in 2009 and homicide totals stayed above 1,000 for most of the following fifteen years, peaking again at 1,616 in 2017 and remaining above 1,300 as recently as 2023, when 1,393 murders were recorded.[6] In 2024, that figure fell to 1,141, a decline of roughly seventeen percent.[6] Then came 2025, in which the Jamaica Constabulary Force recorded 673 murders for the full year, a decline of 43 percent from 2024 and the lowest annual total the country has recorded in 31 years.[1] Police Commissioner Dr Kevin Blake framed the figure as representing more than 470 fewer lives lost compared with the prior year.[7] Shootings fell by roughly 32 percent over the same period and reported rapes declined by around 27 to 28 percent.[8] Early data for 2026 show the decline continuing, with 22 murders recorded islandwide as of mid-January 2026 compared with 40 for the same period the previous year.[9]

This is not a manufactured or cosmetic number. It reflects a real reduction in the number of Jamaicans who were shot, stabbed, or beaten to death in a single year, and it should be read as exactly that.

But a homicide count is a single dimension of a nine-dimension framework, and the moment attention shifts to any other dimension, a very different picture starts to emerge alongside the good news, not in contradiction to it.

Fear as a form of social experience

In March 2025, the Statistical Institute of Jamaica published the 2023 Jamaica National Crime Victimisation Survey, the sixth iteration of a study STATIN conducts roughly every three years on behalf of the Ministry of National Security, based on interviews with 3,294 households conducted between November 2023 and February 2024.[10] Two findings from that survey sit in genuine tension with the homicide data, and both deserve to be taken seriously on their own terms rather than explained away.

First, 86.6 percent of respondents believed crime had increased nationally compared with the 2019 survey period, up from 76.5 percent who believed the same in the previous survey round, and this belief persisted even though 2022 and 2023 had both recorded declines in serious crime according to police statistics.[11] Perception and official trend line moved in opposite directions.

Second, and more specifically, the survey documented a measurable erosion in Jamaicans' sense of safety in certain everyday spaces. Only 68.6 percent of respondents said they felt safe at banks in 2023, down sharply from 80.9 percent in 2019, and only 47.2 percent felt safe using ATMs, down from 56.3 percent previously.[12] Confidence in safety at bars fell from 70.2 percent to 62.8 percent over the same period.[12] Nighttime safety while walking in one's own community also declined, from 69 percent feeling safe at night in 2019 to a lower figure in 2023, even as roughly nine out of ten respondents continued to report feeling safe in the places they visit daily, such as home, church, school, and work.[13]

None of this is a contradiction that needs to be resolved by picking a side. It is two different instruments measuring two different things. Police statistics measure events that were reported and recorded. Victimisation surveys measure lived experience, including the accumulated psychological residue of years in which the murder count sat above 1,000. A society does not update its collective sense of safety the moment a statistic improves. Fear has its own half-life, and Jamaica's 2023 data was collected precisely in the window before the sharpest recent declines had fully registered, which is itself a methodological point worth stating plainly: the next JNCVS round, whenever it is conducted, will be the real test of whether the 2025 homicide drop has actually reached the level of lived experience, or whether Jamaicans continue to organize their days around a risk that official statistics say has receded.

The institutional paradox: falling murders, rising fatal police shootings

Here is where the McEwan framework earns its keep, because the single most striking finding in Jamaica's current security data sits entirely outside the homicide count, and a reader relying on the murder statistic alone would never encounter it at all.

According to the Independent Commission of Investigations, the statutory body responsible for investigating security force conduct, fatal shootings by Jamaica's security forces fell to a low of 86 in 2019, then rose in each subsequent year: 127 in 2021, 134 in 2022, 155 in 2023, and 189 in 2024, a 120 percent increase over the five-year span.[14] In 2025, that figure did not merely continue rising. It surged to 311 fatal police shootings for the year, a year-on-year increase of more than 60 percent and the highest annual total INDECOM has recorded since it began publishing regular statistics in 2017.[15] For thirteen consecutive months beginning in November 2024, Jamaica recorded 20 or more fatal police shootings every single month.[15]

Sit with both numbers at once. In 2025, Jamaica recorded 673 homicides, the lowest total in 31 years. In that same year, Jamaica's security forces killed 311 people, the highest total INDECOM has ever recorded. INDECOM's own reporting has raised specific concerns about the character of these encounters, noting that during planned police operations specifically, the pattern of shootings resulted disproportionately in death rather than injury, and documenting cases in which the conduct of those operations was, in the Commission's own words, questionable.[16] The Commission has also pressed for wider use of body-worn cameras, noting that a large share of fatal encounters occur in operations where no such footage exists.[16]

This is not an argument that the murder reduction is fake, and it is not an argument that every fatal police shooting was unjustified; INDECOM's own reporting acknowledges that fatal shootings represent a small fraction of the total volume of police-citizen encounters, and Commissioner Blake has made the same point publicly.[15] It is an argument that a framework measuring only homicide would have completely missed the fastest-growing form of lethal violence in the country over the past two years. Under the McEwan framework, a sharp rise in fatal encounters between the state and its own citizens, occurring inside an institution with a documented pattern requiring parliamentary oversight, meets several of the ten test criteria at once: a clear pattern rather than an isolated event, a stark power imbalance between an armed state actor and a civilian, a body of institutional and public commentary specifically flagging insufficient accountability mechanisms, and a measurable, worsening trend. Whether it constitutes institutional violence in the fullest Galtungian sense is a judgment this essay leaves open rather than settles, precisely because the framework's discipline requires evidence over assertion. But it plainly belongs in the conversation about violence in Jamaica, and a homicide-only lens would never have brought it there.

What structural indicators show

The structural dimension of the framework tells a genuinely more encouraging story than either the fear data or the police-shooting data, and it deserves equal weight rather than being buried beneath the harder news.

Jamaica's national poverty rate, measured through the Jamaica Survey of Living Conditions and reported by the Planning Institute of Jamaica, fell from 16.7 percent in 2021 to a record 8.2 percent in 2023, the lowest level recorded since the measure began in 1989, before easing further to 7.8 percent in 2024, a figure the PIOJ describes as statistically similar to 2023 given the survey's margin of error.[17] Extreme or food poverty, defined as the inability of a household to afford the minimum daily caloric intake required for good health, fell to 2.8 percent in 2023 and held at roughly the same level, 2.7 percent, in 2024, also a record low.[18] Jamaica's Gini coefficient, the standard international measure of income inequality, improved from 0.3991 in 2021 to 0.3560 in 2023, its lowest level since 2017.[19] PIOJ Director General Dr Wayne Henry attributed the improvement to post-pandemic recovery, expanded employment, and strong remittance inflows, alongside a period of record-low unemployment near 3.7 percent.[19]

Read against Galtung's original concept of structural violence, this matters enormously, and it complicates any narrative that treats Jamaica's violence problem as simply getting worse across every dimension. If structural violence is about systems preventing people from meeting basic needs, then a falling poverty rate and a falling Gini coefficient represent real, measurable movement toward positive peace on that specific axis, even while the institutional and psychological dimensions tell a more troubled story. A serious framework has to be able to hold both of these facts at once without collapsing them into a single verdict.

The economic cost, stated with appropriate caution

The economic literature on the cost of crime in Jamaica is real but should be quoted with care about vintage. A 2017 study by the Inter-American Development Bank estimated that crime cost Jamaica roughly four percent of GDP in 2014, driven substantially by public security spending that exceeded the country's crime-fighting budget for that fiscal year.[20] A follow-up IDB study released in November 2024, using an updated methodology that incorporates human capital losses, private security spending, and public criminal-justice expenditure, found that the direct cost of crime and violence across Latin America and the Caribbean reached 3.44 percent of regional GDP in 2022, and reporting at the time cited Jamaica-specific estimates in the range of five percent of GDP.[21] Regionally, the IDB found that private security spending by businesses and households accounted for the largest single share of these costs, 47 percent, ahead of public spending at 31 percent and lost human capital at 22 percent, and noted that the region's crime bill is equivalent to 78 percent of the entire regional education budget.[21]

These are estimates, not measured accounting figures, and they should be labeled as such rather than treated as precise. What they establish reliably is direction and order of magnitude: violence in Jamaica, as in the wider region, imposes a cost that runs to several percentage points of national output every single year, largely through the diversion of both public and private resources away from productive investment and into protection. That is a real, if imprecisely bounded, form of what this framework calls economic violence, distinct from and additional to the direct physical toll.

Part V: The Jamaican Paradox

Here is the tension this essay has been building toward, and it should not be resolved too quickly or too cleanly, because the honest answer is genuinely uncomfortable.

Jamaica in 2025 and 2026 is, by the most rigorous and reliable measure available, safer than it has been in a generation. Fewer people are being murdered. Fewer people are being shot. Fewer people are being raped, according to reported figures. Poverty has fallen to its lowest recorded level. Income inequality has narrowed. These are not small achievements, and a framework built to interrogate violence has no business pretending they are not real.

And in that same period, the state's own security apparatus killed more civilians than in any year INDECOM has ever recorded, a pattern serious enough that the oversight body specifically flagged operational conduct, disproportionate lethality, and gaps in accountability infrastructure like body cameras. In that same period, a nationally representative survey found that Jamaicans overwhelmingly believed crime was getting worse, not better, and that trust in the safety of ordinary public spaces, banks, ATMs, streets after dark, had measurably eroded compared with four years earlier. In that same period, crime and its associated security spending continued to consume an estimated share of GDP large enough to rival the country's entire education budget.

None of these things is more true than the others. They are simply different dimensions of the same country, measured by different instruments, on different timelines, capturing different forms of harm. A country can become measurably safer by the metric that counts bodies while becoming, by other measures that this essay has tried to make just as rigorous, no less burdened by violence in its broader Galtungian sense. Direct violence and structural violence do not move in lockstep, and there is no law of social physics that requires them to.

This is not a rhetorical trick designed to undercut good news with bad news for dramatic effect. It is what happens when a society finally measures more than one dimension of a genuinely multidimensional problem. The murder count and the fatal-police-shooting count are not in competition to describe "the real Jamaica." They are both the real Jamaica, existing simultaneously, and any account that reports only one of them is not lying, but it is not finished either.

Part VI: Can Violence Exist Without a Body?

Return, for a moment, to the woman walking to the taxi stand at six in the morning.

Under the strict definition most governments use, nothing happened to her. She was not attacked. She does not appear in any crime statistic. If the country's entire understanding of violence rested on that definition, her morning would be invisible, indistinguishable from that of a woman in a city with no crime problem at all who simply prefers a scenic route.

But something did happen, in the sense that matters for a framework built on Galtung's foundation. Her choices were constrained by a threat she had no part in creating. Her potential, in Galtung's own phrase, her freedom to move through her own city without calculation, fell short of what it would otherwise be. That gap between the actual and the potential is the entire definition of violence Galtung proposed in 1969, and it does not require a perpetrator standing in front of her for the gap to be real.

This is the uncomfortable center of the whole exercise. A society can be organized, at the level of daily individual behavior, around violence that is not currently occurring. The taxi route not taken, the phone left at home, the shop closed early, the child kept indoors after school, all of these are rational adaptations to risk, and their cumulative weight across a population is itself a measurable social fact, even when the underlying event they are adapting to never actually takes place on that particular day. The Jamaica National Crime Victimisation Survey is, in a sense, the closest instrument Jamaica currently has for capturing this, and its finding that Jamaicans feel less safe at banks and ATMs than they did in 2019, even as recorded robbery and violent crime trends have shifted, is precisely a measurement of this phenomenon: harm expressed not as an event but as a constraint on ordinary life.

Part VII: From Jamaica to the World

A framework that only works in Jamaica is not a framework. It is a description. The test of the McEwan Index of Social Violence is whether it can be pointed at a society with an entirely different history, geography, and political system and still produce a useful, falsifiable reading.

Consider, briefly, six very different contexts, not as full case studies but as proof of portability.

In El Salvador, the physical-violence dimension has fallen dramatically under an extended state of emergency that has produced one of the steepest homicide declines in the hemisphere. But the institutional-violence dimension of this framework would immediately demand an accounting of mass pretrial detention, due process suspensions, and documented deaths in custody, precisely the kind of trade-off between one form of violence and another that a homicide-only lens would never surface.

In South Africa, the physical-violence dimension registers among the highest homicide rates outside active conflict zones, but the structural-violence dimension, still shaped by apartheid-era spatial and economic segregation decades after its formal end, would likely show the strongest and most durable signal of any dimension in the framework, a textbook illustration of Galtung's original claim that structural harm persists long after the direct violence that built it has ended.

In Finland, homicide and most conventional physical-violence indicators sit near the bottom of global rankings, yet Nordic researchers have documented persistently high rates of intimate partner violence against women relative to the country's overall safety reputation, a case where the physical-violence dimension is genuinely split by gender in a way that a single national homicide rate would flatten and obscure.

In the United States, the framework would need to hold direct violence, structural violence tied to residential segregation, and a uniquely large digital and informational violence dimension, given the scale of documented online harassment and coordinated intimidation campaigns, in view simultaneously, since no single one of them captures the country's full profile.

In Nigeria, physical violence tied to insurgency and communal conflict in some regions would sit alongside a structural-violence dimension shaped by extreme regional disparities in access to electricity, healthcare, and education, illustrating how the same national framework must be applied at a sub-national level in large or federally organized states, since a single national score would erase enormous internal variation.

In Japan, low homicide and low reported violent crime would score well across most of the physical dimension, while researchers studying overwork culture, social isolation among the elderly, and workplace harassment norms have pointed to specific patterns that a cultural or psychological violence lens, applied carefully and without exaggeration, might productively examine.

None of these six sketches is a finished analysis, and none should be mistaken for one. Each is offered only to demonstrate that the same nine dimensions and the same ten-point test can be pointed at radically different societies and generate a different, evidence-specific reading each time, rather than forcing every country into a single Jamaica-shaped template.

Part VIII: Should There Be a Score?

Given a framework this elaborate, the obvious next question is whether it should ultimately produce a single number, a McEwan Violence Index score that could rank countries the way the Global Peace Index or the Human Development Index does.

The honest answer, after building out all nine dimensions, is no, not yet, and possibly not ever in the form of a single composite figure. A single number invites false precision. It would require assigning fixed weights to physical, structural, institutional, economic, psychological, cultural, environmental, temporal, and digital harm, weights that have no defensible empirical basis and that would vary legitimately by context in ways a fixed formula cannot accommodate. It would also invite exactly the kind of crude cross-country ranking, "Country X is more violent than Country Y," that this entire essay has argued against, since it would collapse nine genuinely different phenomena into a comparison that erases the very distinctions the framework exists to preserve.

A more defensible alternative is a multidimensional dashboard rather than a single score: nine separate indicator panels, each populated with the best available data for a given country in a given year, each explicitly dated, each flagged where data quality is weak or absent, with no attempt to compress them into one final digit. Where comparison across countries is genuinely useful, it should happen dimension by dimension, using the McEwan Violence Test as a qualitative check on whether the underlying indicator actually reflects avoidable, patterned, disproportionate harm rather than simply reflecting a raw statistic. This is a less satisfying answer than a ranking table, and it is offered anyway, because intellectual honesty about what this data can and cannot support matters more than the branding appeal of a single headline number.

Part IX: What This Framework Cannot Tell Us

Any framework this ambitious needs an equally honest accounting of its limits, and skipping this section would make the rest of the essay weaker, not stronger.

Correlation is not causation. A society can show both high structural-violence indicators and high physical-violence indicators without one demonstrably causing the other; establishing causal pathways between poverty and homicide, for instance, remains genuinely contested in the criminological literature, and this essay has not attempted to resolve that debate.

Reported crime is not all crime. Every official statistic cited above, from JCF homicide counts to INDECOM shooting data, reflects what was reported and recorded, not the full universe of what occurred, and unreported violence, particularly domestic and sexual violence, is notoriously difficult to capture through official channels alone.

Perception is not identical to objective victimization. The JNCVS finding that most Jamaicans believe crime is rising, even as several official indicators fell over the same period, is real and important, but it measures a psychological and social fact, not a direct contradiction of the police data; treating the two as measuring the same thing would be a methodological error.

Poverty is not automatically violence, inequality is not automatically violence, and institutional failure is not automatically violence. This essay has tried to hold that line throughout, but it bears repeating explicitly, because the entire credibility of the framework depends on resisting the temptation to relabel every form of disadvantage as an act of aggression.

Cultural difference is not automatically cultural violence. Naming a specific belief that legitimizes a specific harm is analytically different from characterizing an entire culture, religion, or community as inherently violent, and this framework should never be used for the latter.

Cross-country comparison is inherently risky. Definitions of crime differ by jurisdiction, survey methodologies differ, and official statistics are revised over time, sometimes substantially, as PIOJ itself has noted regarding pre- and post-2018 JSLC data.

Some violence is fundamentally difficult to observe. Temporal violence, coercive control inside private households, and much of the digital-violence dimension resist the kind of clean empirical measurement this framework otherwise demands, and honesty requires saying so rather than manufacturing indicators that look more solid than the underlying evidence actually is.

Part X: Learning to See

None of this is an argument against celebrating what Jamaica has genuinely achieved. A 43 percent reduction in murders in a single year, four hundred and seventy-some lives that would otherwise have been lost, is not a statistic to be diminished by a longer academic list of everything else still unresolved. It should be stated plainly and without qualification: fewer Jamaicans are being killed than at any point in three decades, and that is a real and significant good.

But a society that only knows how to count bodies will always be surprised by the forms of harm that do not leave one behind. It will not notice the year fatal encounters with the state's own security forces quadrupled from their 2019 low while the murder rate collapsed. It will not notice that a majority of its own citizens have grown less confident walking their own streets after dark, even in years when the official statistics say they should feel safer. It will not notice the quiet arithmetic performed every morning by a woman choosing which route keeps her a little safer, an arithmetic that never shows up in any report because nothing, officially, happened to her at all.

Johan Galtung spent his career arguing that peace is not simply the absence of war, and that violence is not simply the presence of a weapon. Fifty-six years after his first paper on the subject, the argument has lost none of its force, and Jamaica, precisely because it is living through a moment of genuine and measurable improvement on one axis while facing genuine and measurable strain on several others, is as good a laboratory as exists anywhere in the world for testing whether that argument still holds. It does.

A society cannot build a lasting peace around the one thing it has learned to count while remaining blind to everything it has not. Peace, in the fullest sense Galtung meant it, begins the moment a society stops asking only how many people were killed, and starts asking, with equal seriousness, everywhere its citizens have quietly learned to organize their lives around harm that never had to be named to be real.

Sources and Further Reading

[1] Jamaica Constabulary Force / Jamaica Information Service, "Murders Continue Downward Trend" and Ministry of National Security year-end figures, January 2026. [2] Johan Galtung, "Violence, Peace, and Peace Research," Journal of Peace Research, Vol. 6, No. 3 (1969), pp. 167-191. https://doi.org/10.1177/002234336900600301 [3] Galtung (1969), as summarized in academic reviews of the original text, including analysis published via James Madison University Scholarly Commons and the PolSci Institute's typology review. [4] Johan Galtung, "Cultural Violence," Journal of Peace Research, Vol. 27, No. 3 (August 1990), pp. 291-305. https://doi.org/10.1177/0022343390027003005 [5] Galtung (1990), abstract and text as indexed by Oxford Academic / SAGE Journals. [6] Jamaica Constabulary Force annual crime statistics, 2009-2024, as compiled and cross-referenced against UNODC intentional homicide data. [7] Jamaica Information Service, "Hundreds of Lives Saved in 2025 – Police Commissioner," December 30, 2025. [8] Jamaica Information Service, "Murders Continue Downward Trend," January 2026; OSAC Jamaica Country Security Report, February 2026. [9] Jamaica Gleaner, "Jamaica records sharp drop in murders in first weeks of 2026," January 19, 2026. [10] Statistical Institute of Jamaica (STATIN) and Ministry of National Security, 2023 Jamaica National Crime Victimisation Survey (JNCVS) Report, launched March 28, 2025. [11] RJR News / Radio Jamaica News Online, "STATIN Survey Showed More Negative View Of Crime Situation In 2023 Than Police Statistics Indicated," March 28, 2025. [12] Jamaica Gleaner, "STATIN: Jamaicans feel less safe at banks, ATMs, in communities," March 31, 2025. [13] Nationwide 90FM, "Survey: Majority of Jamaicans Feel Safe at Home, Work & Church," March 2025. [14] Independent Commission of Investigations (INDECOM), 2024 Year in Review, cited via Jamaica Observer, "Indecom claims 120% increase in fatal shootings by police since 2019," April 2, 2025. [15] Jamaica Gleaner, "2025: The policing year written in blood," January 4, 2026. [16] Jamaica Gleaner editorial, "Embrace INDECOM's report," October 10, 2025; Nationwide 90FM, "INDECOM Concerned About Year on Year Increase in Police Killings," April 2025. [17] Planning Institute of Jamaica (PIOJ), Jamaica Survey of Living Conditions 2023 and 2024 poverty estimates, as reported at PIOJ quarterly briefings, May and November 2025; Jamaica Information Service, "Poverty at Record Low of 7.8 Per Cent." [18] Jamaica Gleaner, "PIOJ reports decline in poverty," May 22, 2025; Jamaica Information Service, November 2025. [19] Jamaica Observer, "Record drop in poverty," May 22, 2025. [20] Inter-American Development Bank, "The Costs of Crime and Violence: New Evidence and Insights in Latin America and the Caribbean," 2017, as reported in Jamaica Observer, "Crime costs $61.7b," February 2017. [21] Inter-American Development Bank, "The Costs of Crime and Violence: Expansion and Update of Estimates for Latin America and the Caribbean," November 2024; Radio Jamaica News Online, "Crime Hindering Economic Growth In Latin America And Caribbean," November 2024.

Data current as of publication date. Readers are encouraged to verify the latest official JCF, STATIN, PIOJ, and INDECOM statistics, which are updated regularly and may have changed since this analysis was prepared.