Alexander Technique

F. M. Alexander and the early teaching

Frederick Matthias Alexander was born on 20 January 1869 at Table Cape in Tasmania and died in London on 10 October 1955. A University of Tasmania reference entry says he moved to Melbourne to pursue a stage career and that the technique evolved from his efforts to deal with his own vocal problems. The Society of Teachers of the Alexander Technique (STAT) gives an account in which he observed himself in mirrors while reciting and noticed his neck stiffening, his head pulling back and down, and his larynx depressing. That account comes from the teaching body, and it should not be read as established physiology.

Sources differ on when teaching began. STAT says about 1894 in Melbourne, the Australian Dictionary of Biography says he was teaching breathing and voice production by the mid-1890s and worked in Sydney from 1899 or 1900 to 1904, and the University of Tasmania entry says he was teaching in Sydney by 1900. He moved to London in 1904 and taught there and in America for the rest of his life.

Writings, training and institutions

Alexander wrote four major books, including Man's Supreme Inheritance (London: Methuen, 1910, revised in 1918) and The Use of the Self (1932). STAT and the Alexander Trust date the start of his formal three-year teacher training course to 1931, while the American Society for the Alexander Technique (AmSAT) says he began training teachers in London in 1932. STAT says he trained about eighty teachers, a figure that comes from STAT alone. STAT was founded in the UK in 1958 by teachers he had trained and now reports over 600 teaching members worldwide.

The University of Tasmania entry says the technique became established in Australia and more than twenty other countries, and that Alexander had both supporters, among them George Bernard Shaw and John Dewey, and sceptics and detractors.

Reading the founder's writings critically

STAT and the Alexander Trust state that they stand apart from prejudicial or discriminatory passages in Alexander's writings. AmSAT's training directors go further and say that parts of the writings contain disparaging ideas about people of colour, stereotyped ideas of gender and pseudoscientific theories of race, which they reject and condemn. The 1918 edition of Man's Supreme Inheritance includes a chapter titled 'Race Culture and the Training of Children'. How these passages relate to the method as taught today is a matter for specialist historians, and the sources used here do not settle it.

Lessons, teachers and pupils

The teaching profession and the trials of the technique use the words 'lessons' and 'teachers', not 'treatments' and 'therapists'. Cochrane Australia describes the technique as an educational approach in which verbal instruction, gentle hand contact and feedback guide subtle changes in movement, aimed at habitual patterns that proponents suggest may underlie common musculoskeletal conditions. STAT teaching members are registered as MSTAT after an approved three-year training course or an equivalent standard approved by STAT.

Alexander's vocabulary

Within the tradition, 'primary control' refers to the relationship between head, neck and torso, 'inhibition' to pausing a habitual reaction before responding differently, 'direction' to mental instructions such as letting the neck be free, and 'end-gaining' to fixing on the goal of an action without attending to how it is done. These definitions are paraphrased from a small qualitative study of eight participants and are concepts of the Alexander tradition, not measured biomedical constructs.

In his 1918 revised book Alexander wrote of the 'end-gaining principle', holding the end in view and not the 'means whereby', and gave a chapter to inhibition; the phrase 'primary control' does not appear in that book. A teacher-authored article hosted by STAT says the term first appeared in his writing in 1925 and that teachers disagree about its meaning. 'Use', as in The Use of the Self, is Alexander's word for habitual patterns of coordination, but his 1932 text was not open for this draft, so no definition from it has been verified.

Individual lessons

STAT says the technique is learned mainly through one-to-one lessons with a qualified teacher, and that some teachers also offer introductory group classes and workshops. A lesson lasts about 30 to 45 minutes, involves active participation without strenuous exercise, and is taught in unrestrictive clothing. The teacher guides the pupil through everyday movements such as sitting, standing, walking or bending, using verbal explanation and hands-on guidance, and part of a lesson may be spent lying in a semi-supine position.

STAT suggests 20 to 30 lessons as a foundation. This is guidance from the teaching body and not a clinical protocol. Because hands-on guidance is part of most lessons, a person can ask what contact to expect and can ask the teacher to stop.

Settings and delivery in the UK

STAT lists settings such as Alexander centres, clinics, pain clinics, schools, colleges, businesses and teachers' homes. A 2015 national survey of 534 UK teachers from three associations, a 61% response, found that 87% taught on their own premises or in a privately rented room and 19% offered home visits. The authors estimated about 400,000 lessons a year, and 91% of pupils paid privately.

Lesson doses used in trials

In the ATEAM trial, 24 lessons meant 22 lessons over five months, starting at two a week for six weeks, then weekly, then fortnightly, with revision lessons at seven and nine months. Six lessons meant two a week for two weeks and then weekly for two weeks. A later neck pain trial used 20 one-to-one lessons totalling 600 minutes.

Other formats have been tested: a 2025 development study used ten lessons, six in a group and four individual, over 12 weeks and noted that group delivery is not typical in the profession, and an Iranian trial used three 60-minute sessions a week for 12 weeks. These are trial protocols and not standard practice.

Reasons people take lessons

In the UK teacher survey, teachers said that 62% of pupils began lessons for musculoskeletal reasons such as back symptoms, posture, neck pain or shoulder pain, 18% for general reasons including wellbeing, 10% for performance-related reasons such as voice, music or sport, 5% for psychological and 3% for neurological reasons. These reasons were reported by teachers, not by pupils. Pupils were mostly female (66%) with a median age of 48.

Official guidance that mentions the technique

NICE guideline NG71 on Parkinson's disease (2017) recommends that clinicians consider the Alexander Technique for people with balance or motor function problems, after recommendations on physiotherapy. 'Consider' is a weaker recommendation than 'offer'. The NHS says it may offer lessons to people with Parkinson's disease whose balance or movement problems put them at risk of a fall. NICE guideline NG59 on low back pain does not mention the technique, which should not be read as a recommendation for or against it.

Uses studied beyond pain

A systematic review of the technique for musicians found twelve studies, five of them randomised, and reported some evidence that lessons may reduce performance anxiety but inconclusive effects on music performance, respiratory function and posture. A 2012 systematic review of medical and health-related conditions reported only preliminary evidence, insufficient for recommendations, for balance in older adults, general chronic pain, posture, respiratory function and stuttering.

Practitioners have claimed benefits for breathing and asthma. A Cochrane review of the technique for chronic stable asthma found no randomised trials that met its inclusion criteria, which means the claim is untested in that review. That is not the same as showing that the technique does not work.

Chronic and recurrent low back pain: the ATEAM trial

The ATEAM trial randomised 579 adults with chronic or recurrent low back pain from 64 English general practices to normal care, massage, six Alexander lessons or 24 lessons; half of each group also received an exercise prescription with nurse counselling. At 12 months the Roland Morris disability score, which counts activities limited by back pain, was lower than with normal care (control mean 8.07) by 3.40 points after 24 lessons (95% confidence interval -4.76 to -2.03) and by 1.40 points after six lessons. Massage made no significant difference (0.58 points, -1.94 to 0.77).

The median number of days with back pain in the previous four weeks was 21 under normal care and about 18 days fewer after 24 lessons. Outcomes were self-reported by postal questionnaire, and 80% of participants completed the 12-month follow-up.

Neck pain and Parkinson's disease

A UK trial randomised 517 people with chronic non-specific neck pain to 20 Alexander lessons, 12 acupuncture sessions or usual care. At 12 months, neck pain and disability scores were lower than with usual care by 3.79 percentage points with lessons (95% confidence interval 0.91 to 6.66) and 3.92 with acupuncture. A smaller German trial of 72 people found no difference in pain at five weeks between the technique and local heat. These trials differ in comparator, dose and follow-up, so the evidence for neck pain is mixed.

A 2002 trial of 93 people with idiopathic Parkinson's disease compared 24 lessons, 24 massage sessions and no additional intervention. The lessons group improved on a self-assessed disability scale compared with no additional intervention, and the difference was maintained at six months.

Systematic reviews

A 2003 review found four controlled trials, only two methodologically sound, and called the evidence encouraging but not convincing. A 2012 review searched eight databases to July 2011 and included 18 publications, only three of them randomised trials. Its authors graded the evidence as strong for chronic back pain and moderate for disability in Parkinson's disease; those grades rest on few trials.

A 2024 review by Cochrane Australia for the Australian Government (searches to April 2023; seven randomised trials) found moderate-certainty evidence that lessons probably improve physical function in chronic low back or neck pain (three trials, 611 participants) and low-certainty evidence that they may reduce pain. Effects on quality of life, falls and emotional wellbeing were low-certainty or very uncertain. It found no trials for stress, anxiety or mood disorders, acute pain, headache or respiratory conditions, and safety was outside its scope.

Design limits of the trials

The main trials compared lessons with usual care or other treatments, used no sham comparator and relied largely on self-reported outcomes. In ATEAM only data entry was blinded, and participants were told at recruitment that there was suggestive preliminary evidence for each intervention. The protocol of the neck pain trial stated that there was no blinding to the intervention. Participants cannot be unaware that they are having lessons.

Cochrane Australia judged the two largest back and neck pain trials at low risk of bias for several outcomes but raised concern that study limitations may have overestimated benefit for some outcomes, and that selective non-reporting of unfavourable results could have influenced the findings.

Funding and competing interests

ATEAM was funded by the UK Medical Research Council and its authors declared no competing interests. The neck pain trial was funded by Arthritis Research UK. Its paper reports that two authors are members of STAT and two are members of the British Acupuncture Council, the comparator treatment, and the first author of the 2012 review discloses that she teaches the technique privately and sits on STAT's research committee. These disclosures do not show bias, but the evidence is not independent of the teaching profession.

Cost and what remains unknown

In the neck pain trial's economic evaluation, lessons cost GBP 25,101 per quality-adjusted life year gained against usual care. The authors judged them not cost-effective at a GBP 20,000 threshold but possibly so at GBP 30,000, with considerable uncertainty and a high level of missing data. Cochrane Australia reported that longer-term effects were generally not reported. A PubMed search for 2012 to 2026 found no other systematic review specific to Parkinson's disease, voice, performance anxiety or posture, although other databases were not searched.

Adverse events in trials

In ATEAM one participant reported that back pain was made considerably worse by massage, and no adverse events were reported for Alexander lessons or exercise. In the UK neck pain trial no serious adverse event was judged probably or definitely related to lessons or acupuncture. In the German trial, adverse events consisted mainly of slightly increased pain and muscle soreness. The methods for recording adverse events differed between trials, and an absence of reports does not show that none occurred.

Suitability and the limits of the evidence

The trials excluded several groups, so their results cannot be applied to people with serious spinal disease, nerve-root pain, recent spinal surgery or severe limits on walking. No source reviewed for this draft lists formal contraindications. The NHS page that mentions the technique advises telling any therapist about medical conditions and medicines, and asking a GP about symptoms that do not go away. Lessons are not a diagnosis, and new or worsening pain, numbness, weakness or other neurological symptoms are reasons to seek medical assessment.

Training and regulation

The Professional Standards Authority describes Alexander Technique teaching in the UK as an unregulated occupation, and the Complementary and Natural Healthcare Council, an accredited voluntary register, lists it among the professions it registers. STAT states that its teaching members follow a code of professional conduct, are checked through the Disclosure and Barring Service, carry professional indemnity insurance and may also register with that council. These are STAT's own statements and have not been independently audited.

In Ireland the technique does not appear on the regulator CORU's list of regulated professions, which is an inference from the list. In the United States, AmSAT certifies teachers who complete an approved course of at least 1,600 hours over three years; state law was not checked.