A woman walked into my consulting room three months ago and said something I have heard, in some form, from almost every anxiety client I have worked with. “I cannot stop spinning.” She was 45. Two children. Eight years of generalised anxiety, four years of medication, three therapists. The most recent therapist had taught her thought-stopping. She had also tried distraction, journaling, cold showers, podcast recommendations on the science of rumination.
None of it had landed.
When she said spinning, she meant a specific phenomenon. A thought would enter her field around 6 AM. Within twenty minutes it had attached to a worry about her older son, who had a chronic condition. Within an hour it had attached to a worry about her ability to handle the next flare. Within two hours she could not remember what she had walked into the kitchen for. By the time she reached me she would describe her own day as “a fog of nothing.”
This is not a rare presentation. It is not even unusual. It is the most common shape that adult anxiety takes in clinical work in this country.
It also has a name in the Sanskrit literature, and a sequence, written down approximately two thousand years before the modern psychology of rumination was published.
What the Gītā says
In the second chapter of the Bhagavad Gītā, after Arjuna’s collapse and Kṛṣṇa’s first wave of correction, there is a passage that gets passed over in most contemporary treatments of the text. It is verses 62 and 63, chained together.
dhyāyato viṣayān puṃsaḥ saṅgas teṣūpajāyate saṅgāt sañjāyate kāmaḥ kāmāt krodho ‘bhijāyate krodhād bhavati saṃmohaḥ saṃmohāt smṛti-vibhramaḥ smṛti-bhraṃśād buddhi-nāśo buddhi-nāśāt praṇaśyati
Translated literally, holding the technical vocabulary intact:
Contemplating sense-objects, a person develops attachment to them. From attachment, desire is born. From desire, when desire is frustrated, anger is born. From anger comes delusion. From delusion comes the disturbance of memory. From disturbance of memory comes the destruction of intellect. From the destruction of intellect, one is destroyed.
Eight stages, chained.
Read it slowly. This is not a moral teaching. It is a clinical sequence. Each stage produces the next. Each stage is observable. Each stage is, in modern psychological vocabulary, something a clinician can map onto a client’s report. And the sequence runs from a perfectly ordinary mental event — a thought lingering on something — to a state in which the person can no longer think, plan, or remember why they walked into the kitchen.
This is rumination.
Where the modern research caught up
In 1991, Susan Nolen-Hoeksema published a paper in Journal of Abnormal Psychology describing what she called the Response Styles Theory of depression. The argument was that the way a person responds to early signs of low mood predicts whether the mood becomes depression. People who turn toward the bad feeling, rehearse it, examine it, analyse it, end up depressed. People who do not, recover.
She and colleagues subsequently established (Nolen-Hoeksema, 2000; Nolen-Hoeksema et al., 2008) that rumination is a stronger predictor of new-onset major depressive disorder than the original depressive episode itself. Edward Watkins (2008) refined this further, distinguishing constructive from unconstructive repetitive thought.
The neuroscience caught up around the same time. Raichle et al. (2001) identified what they called the Default Mode Network, a set of brain regions that activates during rest and self-referential thinking. Buckner et al. (2008) showed it is the network that builds and rehearses autobiographical narrative. Whitfield-Gabrieli and Ford (2012) and Hamilton et al. (2015) showed that in depression and anxiety, this network is hyperconnected and does not switch off when the person tries to engage in a task.
Marchetti et al. (2016) closed the loop. They showed that the same DMN hyperconnectivity correlates directly with self-reported rumination scores. Whatever the philosophers had been calling the wandering mind, the brain was showing it as a measurable neural signature.
What Nolen-Hoeksema and Watkins and Buckner had separately mapped was, structurally, the first six stages of BG 2.62-63. Object contemplation. Attachment. Desire. Frustration. Confusion. Memory disturbance. The Gītā extends the chain two further stages, to buddhi-nāśa and praṇaśyati — the loss of discrimination and the collapse of functioning. Anyone who has watched untreated anxiety in late adulthood will recognise both.
Why this matters in the consulting room
The clinical leverage of the Gītā version is not that it is older. The leverage is that it identifies the first link.
Most rumination-targeted interventions in the modern literature start at stage three or four. Cognitive restructuring (Beck, 1979) starts when the thought has already taken shape and the affect is already running. Mindfulness-based cognitive therapy (Segal, Williams, & Teasdale, 2002) starts at stage five or six, asking the person to notice the loop without engaging it. These work, sometimes well. Goyal et al.’s 2014 JAMA meta-analysis put effect sizes for mindfulness-based interventions on anxiety at Hedges’ g = 0.38, on depression at g = 0.30.
But for a client who is already in the cascade by the time she sits down, these interventions ask her to halt a process that is most of the way down its own slope.
BG 2.62-63 says the intervention point is stage one.
Dhyāyato viṣayān — when a person is contemplating a sense object. Before attachment forms. Before desire arises. Before frustration produces the cognitive collapse.
This is not abstract. With my 45-year-old client, the protocol was specific. Her cascade started at 6 AM every morning, in bed, before her feet touched the floor. The first contemplation was always the same. Is my son okay today. She did not need to suppress this thought, defuse it, or restructure it. She needed an intervention that came in before the cascade had momentum.
We started with japa. Forty-five seconds of audible Om Namaḥ Śivāya before she opened her eyes. This is not a religious instruction. It is a substitution at the level of dhyāyato viṣayān: the mind contemplates a single sound rather than a sense-object that is about to grow attachment. The breath structure that japa imposes (slow expiratory phase, mandatory pause) doubles as the first half of a pranayama dose (Brown & Gerbarg, 2005).
Within two weeks she could complete the morning sequence without the cascade forming. Within five weeks the cascade had moved from daily to roughly twice weekly, and when it occurred she could interrupt it before stage three. By session ten the GAD-7 was 6, down from 17 at intake.
The framework is what made this faster than thought-stopping or distraction. Distraction, in the Gītā’s terms, is a stage-one re-contemplation: you swap one sense-object for another, but you are still in stage one of the cascade, still vulnerable to attachment, still producing the next link in the chain. Japa is something else. It is a substitution that does not feed the cascade.
What this adds to existing care, not what it replaces
This post is not an argument that the Gītā’s framework outperforms cognitive-behavioural treatments. The DSM is correct that this woman had GAD. The medication was reasonable. The thought-stopping she had been taught was an evidence-based intervention with documented effect sizes (Hofmann et al., 2012).
What the Gītā adds is a stage-one entry point that current rumination treatments largely do not name. It also adds clinical vocabulary. Saṅga, the attachment that follows contemplation, is a richer construct than cognitive fusion. Smṛti-vibhrama, the disturbance of memory in stage six, predicts the working-memory deficits that show up on neuropsychological testing in chronic GAD (Eysenck et al., 2007). The two literatures, ancient and modern, describe the same phenomenon and would benefit from being read together.
For clients, the practical translation is shorter. There is a 2,000-year-old sequence that says rumination begins with a contemplation that has not yet become an attachment. The intervention is to act before that boundary is crossed. With a structured pre-dawn substitution, supported pranayama, and a clear naming of which stage the client is at when the cascade returns, the work moves faster than it does when each instance is treated as a fresh symptom.
That woman is now at session fifteen. The morning sequence has held for sixty-one days. She has stopped describing her own day as a fog.
