Music for grief: what the research shows, and where it stops

Music for grief is something many people reach for, and the research on it is small. In a survey of 772 people, sad music most often brought nostalgia, not sadness. The studies of music with bereaved people hold 7 to 26 people each, and the three that compared groups found grief scores fell further with music therapy. No study has tested one tone for grief.

Ready to play Connection And Communication 639 Hz

Make time to listen before you reply.

Pure 639 Hz tone Open in Studio

Open the listening page

Sixteen long green ribbons of light across dark space, their swing and brightness rising and settling three times from left to right — Illustration, drawn in code. Sixteen ribbons in the green Harmonance gives 639 Hz, the Grieving page's first tone: their swing and brightness rise and settle three times across the picture, the movement back and forth that the dual process model, below, describes in grief.

Why do people turn to music when they grieve?

Often for memory and for company. In an online survey of 772 people, Taruffi and Koelsch asked when and why people choose sad music. Emotional distress was the situation named most often, and among its examples "when grieving for a loss" came up 51 times. In most cases, the authors write, the distress was about a lost relationship.

The same people chose the emotions sad music brings them, as many as applied. Nostalgia came first, at 76 per cent, ahead of peacefulness at 57.5 per cent and tenderness at 51.6 per cent. Sadness itself came lower, at 44.9 per cent. Asked how sad music stirs sadness at all, they rated memory as the most important way. Music that brings someone back is, for many listeners, the point of it.

Interviews with bereaved people say much the same. O'Callaghan and colleagues talked with 8 caregivers after the death of the person they had looked after. Music often lifted their mood and sometimes helped them face their grief, and some kept away from particular music because of the sadness it brought. Their musical lives kept a connection with the person who had died, which mostly supported them and occasionally made things harder. The authors suggest that continuing bonds, the idea that grieving people stay connected to the person through memory, can take in musical memories too: familiar music given new meaning, and new music found that is linked to the person.

What do studies of music and grief actually find?

Small, encouraging results that cannot yet settle much. Six studies of music with bereaved people say how many took part, and together they hold 76 people. Three of them set a music group against people without one and measured grief before and after.

Six bars of how many took part in each study of music with bereaved people: 26, 18 and 10 in the three that compared groups, 8, 7 and 7 in interviews — The studies of music with bereaved people that report their size hold between 7 and 26 people each, 76 in all, and only three compared a music group with people who had none. From each study's own record, and for Young and Pringle from Gillespie and colleagues' 2024 review.

Six studies of music with bereaved people
Study	Who took part	How
Hilliard, 2007	26 children	Three schools, each given music therapy groups, social work groups or a wait, at random
Hilliard, 2001	18 children	Eight music therapy group sessions, or none
Iliya, 2015	10 adults	Music therapy added to usual care, or usual care alone, at random
O'Callaghan, 2013	8 bereaved caregivers	Interviews after a death
Lindenfelser, 2008	7 bereaved parents	Interviews after a child's death
Young and Pringle, 2018	7 family carers	Interviews after a hospice singing group

In Hilliard's 2001 pilot, 18 grieving children had eight sessions of group music therapy or none. The music group's scores moved on a bereavement questionnaire their parents filled in and on a measure of behaviour at home, while their teachers' ratings and their own were less conclusive. In his 2007 study, 26 children at three schools had an hour a week for eight weeks of music therapy groups, social work groups or nothing yet, each school assigned at random. The music therapy group improved on both grief and behaviour, the social work group on behaviour only, and the waiting group on neither.

In Iliya's 2015 pilot, 10 adults living with complicated grief and mental illness were given, at random, grief-focused music therapy on top of their usual care, or usual care alone. Grief scores on the revised Inventory of Complicated Grief fell further in the music therapy group.

The other studies listened rather than measured. Lindenfelser and colleagues interviewed 7 parents whose child had music therapy before dying, and for them it became part of how they remembered. Young and Pringle's 7 family carers, in a hospice singing group after a death, described songs that validated their grief, brought back memories and let feelings out, as Gillespie and colleagues summarise them. With music therapy groups for bereaved teenagers, McFerran and colleagues found no change in how the teenagers saw themselves, practically significant results on coping, and a sample too small to generalise from.

Reviews of the wider evidence agree that it is thin. Gillespie and colleagues gathered 34 studies of music therapy for people caring for someone with a life-threatening illness, before and after the death: 17 qualitative, 8 quasi-experimental, 7 mixed-methods and 2 randomised trials. Too few followed carers past the death for any conclusion about bereavement, they write, and the studies that measured were small and at high risk of bias. For bereaved children and teenagers, a 2010 meta-analysis by Rosner, Kruse and Hagl pooled 27 studies of support, found small to moderate effects, and named music therapy among the promising approaches.

Why does a comparison group matter so much with grief?

Because grief usually softens with time on its own. The American Psychiatric Association notes that for most people grief begins to lessen over time, and the NHS says the same. So a study that measures people only before and after music will see some easing whatever was played.

A 2024 meta-analysis by Hanauer and colleagues, of 39 studies with 5,578 bereaved children and teenagers, shows how large that trap is. For support offered to every bereaved young person, the change in grief was 0.29 standard deviations in studies with no comparison group, and 0.18 in studies with one: almost two fifths smaller once time was allowed for. In the authors' words, "the effects observed in uncontrolled studies are substantially reduced when controlling for the natural course of bereavement".

The three music studies that compared groups have that strength. Their weakness is size: 54 people between them, and in Hilliard's 2007 study whole schools, not single children, were assigned at random. The US National Center for Complementary and Integrative Health says of music research in general that much of it is preliminary, and that few definite conclusions have been reached.

How does grief move, and where can music fit?

One widely used picture of grief is the dual process model of Stroebe and Schut, from 1999. It sees two kinds of task. Loss-oriented tasks face the death itself; restoration-oriented tasks face the life that follows it. A grieving person moves between the two, at times facing and at times setting aside each, and the model counts time off from both as part of coping well. Its authors call the movement oscillation.

The dual process model of grief, with the share of 91 widowed people whose interviews described each part: doing new things was the most common, at 80 per cent, and 79 of the 91 described both sides. Percentages as Bennett, Gibbons and Mackenzie-Smith printed them in 2010.

Parts of the model, and how many described each
Part	Side	Of 91 people
Grief work	Loss	23 per cent
Grief breaking in	Loss	59 per cent
Breaking, keeping or moving bonds	Loss	67 per cent
Avoiding the changes life asks	Loss	27 per cent
Attending to life changes	Restoration	71 per cent
Doing new things	Restoration	80 per cent
Distraction from grief	Restoration	24 per cent
Avoiding or denying grief	Restoration	63 per cent
New roles and relationships	Restoration	54 per cent

Bennett, Gibbons and Mackenzie-Smith read interviews with 91 widowed people aged 55 to 95 in Merseyside, widowed for anything from three months to 32 years, and counted how many described each part of the model. In all, 79 of the 91, or 87 per cent, described both sides, and the authors conclude that oscillation, "or something resembling it", is common among widowed people.

One number in their table does not add up. The paper's own counts for grief work, 17 and 9 people in its two groups, make 26 of 91, which is 29 per cent, while the table prints 23 per cent. The diagram draws the printed figure. Either way, grief work was among the three parts described least.

Music can sit on either side of that movement. The caregivers in O'Callaghan's study used it to face their grief and also to lift their mood, and they turned some music off. One way to read that through the model: a song that brings the person back is loss-oriented work, while music under a new routine, or simply a quiet break, belongs to restoration and rest. Nothing in the research says which a person needs on a given day. The model's point is that both belong.

Can music make grief harder?

It can. Sadness was among the emotions sad music brought for 44.9 per cent of Taruffi and Koelsch's participants, and O'Callaghan's caregivers sometimes kept away from particular music to keep the sadness down. A song tied to the person can also arrive without warning, on the radio or in a shop, and the NHS notes that in grief powerful feelings may appear unexpectedly. Choosing what plays, and stopping it when it is too much, matters more than any particular piece.

A single tone is a different kind of sound. It carries no words and no tune of its own, so it holds no memories until someone gives it one. That makes it unlike a song you shared with the person. Neither is better: they do different jobs.

What does the grief page play?

The Grieving page holds two tones, 639 Hz and the Moon tone at 210.42 Hz, each with a Strings mode. 639 Hz is one of the six Solfeggio numbers, linked in that tradition with connection and relationships, and the 639 Hz guide traces where the number and its meaning came from. The Moon tone is Hans Cousto's octave of the lunar month, set out in the guide to 210.42 Hz. No study has tested either tone with grieving people.

Harmonance measured what the page plays, second by second, from the site's own audio files, with the broadcast loudness meter, EBU R 128. Both plain tones hold almost still: 639 Hz at -22.5 LUFS and the Moon tone at -22.9 LUFS, each within 0.7 LU over three minutes and never moving more than 0.2 LU from one second to the next. Almost all of each tone's energy sits in its one pitch, with its strongest overtone 23.1 dB below it.

Loudness over three minutes of the grief page's two tones and their Strings modes: each tone a nearly level line, each Strings mode with five dips — What the Grieving page plays, measured each second for three minutes: both plain tones hold within 0.7 LU, and both Strings modes average their tone's level with five slow dips at almost the same moments, about every half minute. Measured by Harmonance from the site's own audio files.

The page's four sounds, measured
Sound	Overall, and range	Quietest and loudest second
639 Hz tone	-22.5 LUFS, range 0.7 LU	-22.9 and -22.2 LUFS
639 Hz Strings mode	-22.5 LUFS, range 3.0 LU	-27.7 and -19.6 LUFS
Moon tone	-22.9 LUFS, range 0.7 LU	-23.3 and -22.6 LUFS
Moon Strings mode	-22.9 LUFS, range 3.5 LU	-29.9 and -20.7 LUFS

The Strings modes move as music does. Each averages the same level as its tone, and five times in every three-minute loop the strings thin and dip before they swell back: by 2.9 to 5.2 LU for 639 Hz and by 2.9 to 7.0 LU for the Moon, at almost the same moments, about every half minute. The largest change from one second to the next is 2.4 LU in the 639 Hz strings and 4.0 LU in the Moon's. Among the 639 Hz strings' strongest pitches is 159.68 Hz, less than a cent from 159.75 Hz, two octaves below 639 Hz.

No study of grief tested either sound. What the readings show is what the page offers: a plain tone that stays where it is, or strings that rise and fall slowly, at whatever level you set. The 639 Hz page and the Moon tone's page play each on its own, the Solfeggio collection holds 639 Hz's neighbours, and Harmonance Studio lets you set a timer or lay a tone under ambient sound.

When is grief more than grief?

For most people grief eases with time, though rarely in a straight line. For some it does not. The American Psychiatric Association describes prolonged grief disorder: intense longing for the person, or preoccupation with their death, that persists and interferes with daily life, at least a year after the death for adults and six months for children and teenagers. It estimates that 4 to 15 per cent of bereaved adults experience it, and that therapies using elements of cognitive behavioural therapy have been found effective. The NHS suggests seeing a GP when feelings do not get easier as the months go by, or when a low mood has lasted more than 2 weeks.

Support is free. In the UK, Cruse Bereavement Support runs a helpline on 0808 808 1677, answered by trained bereavement volunteers. If grief brings thoughts of not wanting to live, in the UK you can call NHS 111 or Samaritans on 116 123, as the NHS urgent help page lists, and in the US you can call or text 988, the 988 Suicide and Crisis Lifeline.

How might you try music while grieving?

Choose by the day. Some days a song that holds the person may be what you want; on others, music with no memories in it, or silence. The dual process model makes room for both, and for rest from both.
Keep the off switch close. If something hurts more than it helps, stop it. O'Callaghan's caregivers set some music aside and still used other music to lift their mood.
Keep it low. A steady sound in the background needs very little volume: just enough to hear.
Share it, if you like. Most of the studies above met in groups, from children's music therapy groups to a hospice singing group for carers, and the NHS suggests peer support, people using their own experience to help each other.
Know what the studies tested. The music in them was led by trained music therapists, which is a different thing from listening alone, and none of them used a single tone.

The Grieving page plays both tones in one place. For more on the research behind listening, the guide to what sound healing research shows looks at the wider evidence, the guide to music and sleep covers a larger body of trials, and the guide to using your own voice is a place to start if singing appeals.

Sources and limits

The research on music and grief is small. The six studies with known numbers hold 76 people between them, the three that compared groups hold 54, and none tested a single tone, 639 Hz, the Moon tone or any Harmonance sound. The survey of sad music asked about sad music in general, not grief alone, and its participants averaged 28 years old. The dual process figures come from older widowed people in one part of England, interviewed on average more than eight years after their loss. The loudness readings come from the site's own audio files, measured by Harmonance: the strings at the page's default level, without the quiet room sound the page adds beneath them. How loud either is in your room depends on your device and its volume.

Tones to play while you read

Just be curious. This is not a medical device or medical advice. For any health concern, please speak to your doctor.