How Was Liberating Insight Related to the Development of the Four Jhānas in Early Buddhism? A New Perspective through an Interdisciplinary Approach

Abstract

The precise nature and status of the meditative states known as the four jhāna-s in early Buddhist soteriology is one of the most controversial subjects of early Buddhist studies. Amongst the most unclear issues connected with jhāna meditation is its relation to liberating insight. There appear to be fundamental discrepancies related to this issue in the Suttapiṭaka itself and in the later Buddhist meditative texts. These discrepancies appear to be sometimes difficult or even impossible to reconcile. In this paper I attempt to present a model of liberating insight as an intrinsic quality of the jhāna meditative state through an interdisciplinary approach relying on textual studies as well as on the new developments in the field of rapidly developing cognitive sciences. In the first part of the paper I analyze various concepts of liberating insight present in the Suttapiṭaka and the way they are connected to the development of the four jhāna-s. Then I point out some fundamental difficulties connected with the traditional Buddhist model of insight understood as a meditative method on its own, distinct from a jhāna meditative state. Later I attempt to propose an explanation of how and why the original concept of liberating insight as an intrinsic quality of jhāna states underwent a radical evolution, which has unfortunately led to both textual discrepancies and serious problems on a practical and psychological level. In order to provide a plausible model of liberating insight as an intrinsic aspect of a jhāna state, I will also refer to some important new developments from the field of cognitive sciences, which provide a new way of explaining how human cognition works. In order to show that my model is possible on a practical level, I will also point out some meditative developments from the later history of Buddhism, where insight was seen in a way somewhat similar to what I am proposing.

Full text

How Was Liberating Insight Related to the Development of the
Four Jhānas in Early Buddhism? A New Perspective through an
Interdisciplinary Approach
Grzegorz Polak
The precise nature and status of the meditative states known as the four
jhāna-s in early Buddhist soteriology is one of the most controversial
subjects of early Buddhist studies. Amongst the most unclear issues
connected with jhāna meditation is its relation to liberating insight.
There appear to be fundamental discrepancies related to this issue in
the Suttapiṭaka itself and in the later Buddhist meditative texts. These
discrepancies appear to be sometimes difcult or even impossible
to reconcile. In this paper I attempt to present a model of liberating
insight as an intrinsic quality of the jhāna meditative state through an
interdisciplinary approach relying on textual studies as well as on the
new developments in the eld of rapidly developing cognitive sciences.
In the rst part of the paper I analyze various concepts of liberating
insight present in the Suttapiṭaka and the way they are connected to the
development of the four jhāna-s. Then I point out some fundamental
difculties connected with the traditional Buddhist model of insight
understood as a meditative method on its own, distinct from a jhāna
meditative state. Later I attempt to propose an explanation of how and
why the original concept of liberating insight as an intrinsic quality of
jhāna states underwent a radical evolution, which has unfortunately led
to both textual discrepancies and serious problems on a practical and
psychological level. In order to provide a plausible model of liberating
insight as an intrinsic aspect of a jhāna state, I will also refer to some
important new developments from the eld of cognitive sciences,
. 6(5): 85–112. ©6 Grzegorz Polak

86
LIBERATING INSIGHT
which provide a new way of explaining how human cognition works.
In order to show that my model is possible on a practical level, I will
also point out some meditative developments from the later history of
Buddhism, where insight was seen in a way somewhat similar to what
I am proposing.
The precise nature and status of jhāna meditation in early Buddhist soteriology
remains one of the most controversial subjects of early Buddhist studies.
Amongst the most unclear issues connected with jhāna meditation is its relation
to liberating insight. The English word “insight” itself is most often used in
modern times as a direct translation of the Pāli term vipassanā (cf. Bodhi, 2000:
330, 397), and occurs very frequently in meditative literature. In the Suttapiṭaka
itself, however, it does not occur very frequently (cf. Rhys Davids & Stede,
2007: 627). It becomes much more prominent in Buddhaghosa’s Visuddhimagga,
which describes the vehicle of pure insight (vipassanāyāna – Vism XVIII.15),
various insight knowledges (vipassanāñāṇāni – Vism XX.104-XXI.61), and
the ten imperfections of insight (dasa vipassanupakkilesā – Vism XX.105). We
nd the term ”liberating insight” in the work of many modern scholars who
discuss early Buddhist meditation. Its meaning seems to be far wider than that
of vipassanā, however. It features prominently in Lambert Schmithausen’s
inuential paper On Some Aspects of Descriptions or Theories of “Liberating
Insight” and “Enlightenment” in Early Buddhism (e.g. Schmithausen, 1981: 199,
204). As Schmithausen has remarked, he was not concerned with all the aspects
of liberating insight, but mainly with the issue of its content (Schmithausen,
1981: 199). He focuses in particular on the insight with regard to the four noble
truths, which is rendered in the Suttapiṭaka by such terms as:
abbhaññāsiṃ, in the autobiographical version, but jānāti in the
versions describing the Path of the Liberation of the Disciple.
Afterwards, however, both versions refer to this comprehension by
means of “jānato … passato” (Schmithausen, 1981: 204).
The term “liberating insight” has also been frequently used by Johannes
Bronkhorst in his seminal work The Two Traditions of Meditation in Ancient
India (e.g. 1986: 96, 97, 102, 104). Bronkhorst (1986: 101) points out that
the Buddhist texts often speak about “insight” (s. prajnā/paññā) as something
immediately preceding liberation and that liberating insight takes place in the
fourth jhāna (Bronkhorst, 1986: 97). He suggests that originally these passages

LIBERATING INSIGHT
87
merely made a short reference to paññā (Bronkhorst, 1986: 102). Elsewhere
(Bronkhorst, 1986: 114) he mentions three insights (ñāṇa-s). Therefore, we may
conclude that when Bronkhorst uses the term “liberating insight” he does so
with reference to the Pāli terms paññā or ñāṇa.
Tilmann Vetter is yet another scholar who refers to the concept of “liberating
insight” in his book, The Ideas and Meditative Practices of Early Buddhism.
He makes an interesting distinction between paññā, which he labels as
“discriminating insight” and aññā, which he considers the “right insight” or
“liberating insight” (Vetter, 1988: 30, 32).
Alexander Wynne’s The Origin of Buddhist Meditation is a more recent work
dealing with the issue of “liberating insight” (e.g. Wynne, 2007: 120-121, 123-
124). He has pointed out that it is possible to nd a notion of non-intellectual
liberating insight in the Posālamāṇacapucchā of the Pārāyanavagga (Sn 1112-
15), which describes a meditator who sees (vipassati) after having grasped
(abhiññāya) and thus becomes liberated1 (cf. Wynne, 2007: 105). Later he
(2007: 120) speaks of a more intellectual form of liberating insight (paññā).
Wynne is probably the rst scholar who has come up with a denition of what
may be considered liberating insight:
Instead of attaining a complete cessation of thought, some sort of
mental activity must take place: a liberating cognition based on the
practice of mindful awareness (Wynne, 2007: 105).
As we have seen, the term ‘liberating insight’ is used by modern scholars
with reference to several Pāli terms (e.g. paññā, aññā, ñāṇadassana,2 abhiññā,
1 Sn 1115: ākiñcaññasambhavaṃ ñatvā, nandī saṃyojanaṃ iti. Evam etaṃ abhiññāya, tato
tattha vipassati.
2 It is noteworthy, that ñāṇa-dassana is a Jain term and in Jainism it functions as a dvandva
compound with the meaning of the two achievements clearly differentiated. I am grateful to
Richard Gombrich for pointing this out to me. This is probably an example of the Buddha’s
tendency to use terms which were already in circulation during his times but to provide them
with a new meaning. It is also worth noticing that the term paññā appears in the Suttapiṭaka
in connection with the pre-Buddhist teachers Āḷāra Kālāma and Rāma (e.g. MN 26/I 164-165),
where it appears as a part of a vefold set together with saddhā, viriya, sati and samādhi. This
set of course also appears throughout the Suttapiṭaka as an element of the Buddha’s own teaching
and is known as the ve indriya-s (e.g. SN 48.1/V 193). But does that mean that the term paññā
was already in usage among the pre-Buddhist sects, perhaps even in the meaning of “liberating
insight”? I believe that we cannot make such a conclusion, as we have no access to Āḷāra Kālāma’s
and Rāma’s own formulations of their teachings.

88
LIBERATING INSIGHT
vipassanā). It appears that it is not the presence of any term in itself, but the
context in which it appears, that decides whether it refers to liberating insight
or not.
Therefore, in this paper, when I refer to “liberating insight”, I understand it
as a cognitive act leading to seeing things as they really are and resulting in a
transformation of a human being and feeling certain of one’s own liberation. This
tentative denition is general enough to leave room for some new interesting
possibilities of understanding liberating insight.
There appear to exist fundamental discrepancies related to the issue of
liberating insight in the Suttapiṭaka itself and in the later Buddhist meditative
texts. These discrepancies appear to be sometimes difcult or even impossible
to reconcile.
As Schmithausen has noted:
There are already in the Sūtra Piṭaka various, even conicting
views or theories of Liberating Insight (and Enlightenment)
(Schmithausen, 1981: 240).
We can indeed nd many different concepts of liberating insight, and in many
but not all of them it is closely connected to the development of the four jhāna-s.
The scheme of liberation in which liberating insight takes place in the fourth
jhāna and is achieved by directing the mind towards the destruction of the
efuents (āsava) has already been given much attention by scholars. It consists
of insight into the four noble truths, and later the same fourfold model is applied
to the āsava-s. To recapitulate: the consensus is that this complicated scheme
cannot be accepted as representing the original account of enlightenment
(Schmithausen, 1981: 205). The knowledge of the four noble truths, which
probably in itself is pretty authentic, appears to have no place in this context.
Schmithausen (1981: 208) has noted the psychological implausibility of insight
into the four noble truths bringing an end to desire. Bronkhorst has on the other
hand stated that the four noble truths are useful knowledge for someone who
is about to enter upon the path leading to liberation, but are long overdue for
someone at the end of the road:
We observed that knowledge of the four Noble Truths must come at
the beginning of the path leading to ‘the cessation of suffering.’ […]
They constitute what an aspirant must know before he can actually
go the path and be liberated (Bronkhorst, 1986: 99).

LIBERATING INSIGHT
89
Bronkhorst’s view harmonizes very well with the position that right view
(sammādiṭṭhi), dened as the knowledge of the four noble truths, occupies in the
scheme of the noble eightfold path. Right view is dened as the knowledge of the
four noble truths in the Vibhaṅga Sutta (SN 45.8/V 8) contained in the Magga
Saṃyutta of the Saṃyutta Nikāya. This sutta explicitly states that right view
(sammādiṭṭhi) is the knowledge (ñāṇa) of suffering, of the origin of suffering, of
the cessation of suffering and of the way leading to the cessation of suffering.3
This is probably the most “classical” denition of right view.
I believe that there are good reasons to suppose that, at least at some point,
the noble eightfold path may have represented a set of factors which had to be
realized in a gradual way. The order in which these factors appear in the noble
eightfold path would therefore correspond to the order in which they should be
developed. The development of the rst factors, such as sammādiṭṭhi, would be
a necessary condition for the development of the factors that follow it in the set.
Support for such an interpretation may be found in some of the suttas
contained in the Dasaka Nipāta of the Aṅguttara Nikāya. Several suttas (from
AN 10.103/V 211 to AN 10.112/V 222) mention a set of ten factors, in which
the elements of the noble eightfold path are followed by sammāñāṇa (right
knowledge) and sammāvimutti (right release). The Vijjā Sutta (AN 10.105/V
214) appears to be particularly interesting in our case. It states that for one of
right view (sammādiṭṭhi kassa) there arises right intention (sammāsaṅkappo).4
A relation of a similar kind is mentioned for each of the following factors, and
the acquisition of the preceding factor appears to be a necessary (and almost in
fact a sufcient) condition for the arising of the next factor. On the other hand,
in the earlier part of the sutta we read that for one of wrong view (micchā diṭṭhi-
kassa) there is also wrong intention (micchāsaṅkappo).5 The presence of wrong
intention causes in turn the arising of the other factors of the set in their "wrong
version". If we are to take the message of this sutta seriously, it would imply
that one simply cannot properly develop any of the later factors without having
rst developed right view. On the other hand, the acquisition of right view must
be seen as a necessary condition for the further development of the factors that
follow it in the set. If we accept the denition of right view contained in the
3 SN 45.8/V 8: Katamā ca, bhikkhave, sammādiṭṭhi? Yaṃ kho, bhikkhave, dukkhe ñāṇaṃ,
dukkhasamudaye ñāṇaṃ, dukkhanirodhe ñāṇaṃ, duk kha nirodha gāminiyā paṭipadāya ñāṇaṃ—
ayaṃ vuccati, bhikkhave, sammādiṭṭhi.
4 AN 10.105/V 214: sammā diṭṭhi kassa sammāsaṅkappo pahoti.
5 AN 10.105/V 214: micchā diṭṭhi kassa micchāsaṅkappo pahoti.

90
LIBERATING INSIGHT
Vibhaṅga Sutta, then it lends support to Bronkhorst’s statements that knowledge
of the four noble truths must come at the beginning of the path leading to the
cessation of suffering.
Schmithausen has pointed out a very important thing: that there is a difference
between liberating insight and the awareness of liberation. The latter may be
understood as feeling certain of being liberated from suffering and having
reached a stage from which one does not fall back. The formulas describing
this certitude of liberation seem to bear the marks of authenticity due to their
simplicity.
There are probably two most noteworthy formulas of that type in the
Suttapiṭaka. One of them is used both in an account of “gradual training” (e.g.
DN 2/I 47) and in some accounts of the Buddha’s own awakening (e.g. MN 85/
II 94). It has the following form:
When liberated (vimuttasmiṃ), there arose a knowledge (ñāṇaṃ
ahosi): “is liberated” (vimuttamiti). I directly knew (abbhaññāsiṃ):
“birth is exhausted (khīṇā jāti), holy life has been lived (vusitaṃ
brahmacariyaṃ), what ought to be done is done (kataṃ karaṇīyaṃ),
there is nothing more for this state (nāparaṃ itthattāya).”6
The second may be found in the account of the Buddha’s own awakening in
the Ariyapariyesanā Sutta (MN 26/I 160):
There arose (udapādi) for me knowledge (ñāṇa) and vision
(dassana)—unshakable is my release (akuppā me vimutti), this
is the last birth (ayam antimā jāti), there is now no more further
becoming (natthi dāni punabbhavo).7
Alexander Wynne has pointed out that this formula of awakening is unique, in
that it contains a pericope that is used throughout the Canon only in connection with
the Buddha’s own awakening (Wynne, 2007: 20). Therefore Schmithausen (1981:
207) has noted that the nal knowledge or awareness of being liberated seems to have
been regarded as an essential element from the very beginning. In these formulas we
nd terms of interest such as ñāṇa, dassana, and an aorist of abhijānāti.
6 MN 85/II 094: vimuttasmiṃ vimuttam iti ñāṇaṃ ahosi. ‘khīṇā jāti, vusitaṃ brahmacariyaṃ,
kataṃ karaṇīyaṃ, nāparaṃ itthattāyā’ ti abbhaññāsiṃ.
7 MN 26/I 167: ñāṇañca pana me dassanaṃ udapādi — ‘akuppā me vimutti, ayam antimā jāti,
natthi dāni punabbhavo’ti.

LIBERATING INSIGHT
91
While these formulas express the certitude of liberation and fullment,
they say nothing at all about the content of liberating truth that was
supposedly discovered. They seem to express the immediate results of
awakening. Apparently, we cannot infer anything from these formulas about
the nature of the cognitive act (i.e. liberating insight) which has resulted
in feeling certain of one’s own liberation. The fourfold scheme of insight
into the āsava-s has been rightly identied as a result of later modications
by Schmithausen (1981: 205), Bronkhorst (1986: 98) and Wynne (2007:
124). This leaves the four jhāna-s, the destruction of the āsava-s (but not
the fourfold scheme) and the certitude of liberation as relatively authentic
elements.
In many of the suttas, we nd a different concept of liberating insight
connected with the practice of the four jhāna-s. In these suttas, the
imperfections connected with the jhānic states become themselves the
object of insight. For example, the Jhāna Sutta (AN 9.36/IV 422) of the
Aṅguttara Nikaya suggests that while absorbed in any of the four jhāna-s,
or the four arūpa-s (formless states) and saññāvedayitanirodha (cessation
of perception and feeling) one regards (samanupassati) whatever element
there is connected to any of the ve khandha-s as impermanent (aniccato),
painful (dukkhato), void (suññato), non-self (anattato), among other labels
(rogato gaṇḍato sallato aghato ābādhato parato palokato) conveying the
painful, dissatisfying nature of existence. As a result, he keeps his mind
back from those phenomena (dhammehi cittaṃ paṭivāreti)8 and “focuses”
(upasaṃharati) on the deathless property (amata dhātu) in the following
way:
‘this is peaceful (santaṃ) this is excellent (paṇītaṃ), namely calming
of all that is constructed/made up (sabbasaṅkhārasamatho),
giving up of all clinging (sabbūpadhipaṭinissaggo), destruction
of craving (taṇhākkhayo), dispassion (virāgo), cessation
8 PTS version has paṭivāpeti, but it seems impossible to provide a plausible etymology of
this term. I am indebted to Richard Gombrich for pointing this out to me. He has also suggested
emending to paṭivāreti, which ts much better in this context. Interestingly, the Thai edition has
patiṭṭhāpeti and the Cambodian one has paṭipādeti. This may perhaps suggest that some corruption
of the original text occurred during the transmission so that there was uncertainty concerning the
verb used in this fragment.

92
LIBERATING INSIGHT
(nirodho) and Nibbāna.9
This in turn, will result either in the destruction of the āsava-s, or in becoming
an opapātika (spontaneously reborn being) and achieving nal release in that
state. A similar concept is proposed in the Aṭṭhakanāgara Sutta (MN 52/I 349)
of the Majjhima Nikāya, although slightly different terms are being used. After
the attainment of any of the four jhāna-s, the meditator discriminates/reects
(paṭisañcikkhati) that the attained jhāna is made up/constructed (abhisankhata),
and intended/planned (abhisañcetayita). Then he understands (pajānāti) that
such a state is impermanent and subject to cessation (nirodhadhamma).10 The
result of such insight is the same as in the aforementioned Jhāna Sutta. The same
form of insight is then applied to any of the nine successive states culminating
in saññāvedayitanirodha, and to the development of loving-kindness (mettā),
compassion (karuṇā), equanimity (upekhā) and sympathetic joy (muditā) as well.
These two suttas differ from the ones analyzed above in that they describe
a different content of liberating insight and a different mechanism by which
liberation occurs. Vision of jhāna as a dissatisfying, conditioned state is the
content of liberating knowledge. This results in disenchantment and in turn in
the destruction of the efuents.
It is easy to see that this concept of insight is based on a new vision of the
four jhāna-s, now no longer considered the central teaching of the Buddha. The
fourth jhāna is no longer seen as a special, puried state – in these suttas it is just
a stage between the third jhāna and the attainment of the base of innite space
(ākāsānañcāyatana). The concept of opapātika present in these suttas also seems
to represent a later stage of development which harmonizes well with the new
vision of the jhāna-s. Apparently at this point nal and irreversible liberation no
longer seemed so certain, possibly due to the growing confusion about the nature of
authentic Buddhist practice. The introduction of the concept of opapātika could shift
the nal liberation to a future existence, and thus provide meaning and hope to the
life of a person who has failed to reach the ultimate goal of Buddhism here and now.
9 AN 9:36/IV 423: so yad eva tattha hoti rūpagataṃ vedanāgataṃ saññāgataṃ saṅkhāragataṃ
viññāṇagataṃ, te dhamme aniccato dukkhato rogato gaṇḍato sallato aghato ābādhato parato
palokato suññato anattato samanupassati. so tehi dhammehi cittaṃ paṭivāpeti. so tehi dhammehi
cittaṃ paṭivāpetvā amatāya dhātuyā cittaṃ upasaṃharati — ‘etaṃ santaṃ etaṃ paṇītaṃ yadidaṃ
sabbasaṅkhārasamatho sabbūpadhipaṭinissaggo taṇhākkhayo virāgo nirodho nibbānan’ti.
10 MN 52/I 350: ‘idampi paṭhamaṃ jhānaṃ abhisaṅkhataṃ abhisañcetayitaṃ. yaṃ kho pana
kiñci abhisaṅkhataṃ abhisañcetayitaṃ tad aniccaṃ nirodhadhamman’ti pajānāti.

LIBERATING INSIGHT
93
Moreover, the method of insight described here seems to be implausible from
a psychological point of view. The fact has been expressed by some Buddhist
scholars, such as Henepola Gunaratana in A Critical Analysis of the Jhānas in
Theravāda Buddhist Meditation:
Insight cannot be practiced while absorbed in jhāna, since insight
meditation requires investigation and observation, which are
impossible when the mind is immersed in one-pointed absorption
(Gunaratana 1985:151).
But if this is indeed the case, wouldn’t it make liberating insight taking
place in the fourth jhāna and leading to the destruction of the āsava-s equally
implausible? One might ask then, why should we single out insight into the
imperfections of the jhāna-s as a later development? There are a couple of
issues to be considered here. If we analyze the passages in the form in which
they have survived into modern times, then indeed fourfold insight into the four
noble truths and the āsava-s taking place in the fourth jhāna is psychologically
implausible. This is in fact yet another argument supporting the relative lateness
of the passage in its present form, in addition to those of Schmithausen (1981,
207-208) or Bronkhorst (1986: 98). However, those scholars have also suggested
that the present form of the account is probably a result of modication of a
more original, authentic account. Bronkhorst has stated:
Let us see what remains that can be considered authentic Buddhist
meditation in view of the conclusions of the present chapter. The
Four Dhyānas and the subsequent destruction of the intoxicants
survive the present analysis easily (Bronkhorst, 1986: 88).
Schmithausen has commented that his issue:
is not with the antiquity of the notion of āsravas as such […]
Therefore it seems preferable to consider the whole “āsrava-layer”
as genuine (Schmithausen, 1981: 206).
The second issue deserving consideration is that insight requiring
investigation and observation, as mentioned by Gunaratana, may not be the only
type of insight. Such a possibility has already been hinted at by Bronkhorst
(1986: 104), who has stated that liberating insight acknowledged by the earliest
Buddhist tradition remained unspecied, or even that it was in fact unspeciable

94
LIBERATING INSIGHT
(Bronkhorst, 1986: 102). Alexander Wynne has commented that the simpler,
non-intellectual versions of liberating insight are likely to be earliest, though
the content of liberating insight in the earliest teaching is unclear (Wynne, 2007:
124). The occurrence of such a type of insight in the fourth jhāna may therefore
not be psychologically implausible at all. The insight mentioned in the Jhāna
Sutta and in the Aṭṭhakanāgara Sutta is however of the traditional type, so the
issue of psychological implausibility still remains.
It seems valuable to compare such canonical concepts with the real life
experience of modern meditators. That is because the later concepts present
in the suttas may often be a result of doctrinal evolution and polemics, and do
not necessarily reect authentic practices and experiences. Although we cannot
be sure that the experiences of modern meditators have any connection with
the ancient texts, it is nevertheless worthwhile considering whether they might
offer a sort of a view from the inside. Such a view might help to explain some
general features of Buddhist meditation, and so clarify the problems and puzzles
which abound in the canonical sources. Ajahn Brahm, a modern meditation
master, writes about jhāna in his Mindfulness, Bliss, and Beyond: A Meditator’s
Handbook:
From the moment of entering a jhāna, one will have no control.
One will be unable to give orders as one normally does. One cannot
even decide when to come out.11 […] Thus in jhāna not only is there
no sense of time but also no comprehension of what is going on
(Brahm, 2006: 153).
No decision making process is available. […] consciousness
is nondual, making comprehension inaccessible […] (Brahm,
2006:155).
In other words, the concept of insight into the imperfections of the jhāna-s
which is supposed to be practised while being at the same time absorbed in
the very state of jhāna might not only be a later development, but might also
be implausible on a psychological level. The Samaṇamaṇḍikā Sutta (MN 78/
11 It is interesting to notice that the problem of emerging from a meditative state devoid of
thought has already been touched upon in the Cūḷavedalla Sutta (MN 44/I 299). The sutta deals
with the emergence from the attainment of cessation of perception and feeling (saññā ve dayi ta-
nirodha samā pattiyā vuṭṭhānaṃ). This text rejects the notion of any decision making process on
part of the meditator, but bases the moment of emergence on the previous development of the mind.

LIBERATING INSIGHT
95
II 22) states that the unskilful intentions (akusalā saṅkappā) cease without
remainder (aparisesā nirujjhanti) in the rst jhāna and the same happens to
the good ones (kusalā saṅkappā) in the second jhāna.12 It is not even possible
to think about starting a different practice while one is absorbed in jhāna. If
one were indeed to start such a practice, it would mean in fact that one was no
longer in the state of jhāna.
The development of this new concept of insight is undoubtedly a result of the
doctrinal evolution of the concept of the jhāna-s themselves. From some point
they were no longer seen as an exclusively Buddhist form of meditation, and
the fourth jhāna had lost its elevated status. But still there must have remained a
belief that liberating insight must be connected with the state of jhāna, and in an
attempt to somehow harmonize this old belief with the new vision of the jhānic
states, the imperfections of the jhāna-s became an object of insight.
In the Visuddhimagga, Buddhaghosa seems to have realized that such a
concept is psychologically implausible. In the method of the vehicle of serenity
(samathayāna), one has to emerge from jhāna and then make this state, now
a thing of the past, the object of one’s insight (Vism XVIII.3). Somehow the
strength of concentration and the freedom from the hindrances (nīvaraṇa)
is supposed to be carried over to the state immediately following the jhāna.
An even bigger problem is connected with the notion of practising insight
with regard to a state that is not in the present, but only in the memory of the
meditator, for this seems to represent a departure from the Buddhist postulate
that insight should be concerned with things as they are (yathābhūtaṃ) in their
directly known form.
Several suttas speak about insight without any mention of the four jhāna-s at
all. In these texts, the meditator reaches liberation as a result of seeing the various
elements constituting his body/mind complex as impermanent (aniccaṃ), and
therefore stressful (dukkhaṃ) and as a result non-self (anattā). That which is
non-self should be seen (daṭṭhabbaṃ) by right understanding (sammappaññāya)
just as it is (yathābhūtaṃ), thus: This is not mine (n’ etaṃ mama), this I am not
(n’ eso ’ham asmi), this is not my self (na m’ eso attā). This leads to weariness/
12 MN 78/II 28: […]paṭhamaṃ jhānaṃ upasampajja viharati; etth’ ete akusalā saṅkappā
aparisesā nirujjhanti.[…] dutiyaṃ jhānaṃ upasampajja viharati; etth’ ete kusalā saṅkappā
aparisesā nirujjhanti.

96
LIBERATING INSIGHT
disenchantment (nibbidā), dispassion (virāga) and liberation (vimutti).13
Several different Buddhist theoretical schemes of body/mind complex may be
used in this context, such as those of the four or six elements (dhātu – e.g. the
Dhātuvibhaṅga Sutta MN 140/III 237), sense bases (āyatana – e.g. the Mahāsaḷ-
āyatanika Sutta MN 149/III 287), khandha-s (Khandha Saṃyutta - SN 22/III 1),
or simply feelings (vedanā – Vedanā Saṃyutta SN 36/IV204).
On a practical level, many of these accounts do not seem to serve well either
as instructions for practice or as verbalizations of the immediate experience
representing the supposed content of the liberating insight. How is one supposed to
directly know and see such elements as “eye” (cakkhuṃ) or “eye-consciousness”
(cakkhuviññāṇaṃ) (SN 35.25/IV 16)? If we are to take Buddhist schemes of
cognitive processes seriously, these elements would represent “subjective/
internal (ajjhatta) conditions (paccaya)” necessary for the arising of experience,
but not the experience itself. It also seems that according to this mode of analysis
we are not in fact aware of visual forms (rūpa) in themselves, as they are the
“objective/external (bahiddhā) conditions” of our experience. It would seem that
the elements which can be directly experienced start with anything that is felt
(vedayitaṃ) as pleasant (sukhaṃ), unpleasant (dukkhaṃ) or neither unpleasant
nor pleasant (adukkhamasukhaṃ) and which arises based on the contact of all the
above mentioned subjective and objective conditions (cakkhusamphassapaccayā
uppajjati).14 The same can be said about the basic elements (dhātu), which either
appear as a set consisting of four elements (catudhātu - paṭhavī, āpo, tejo, vāyo –
SN 14.30/II 169) or six (the above-mentioned four plus ākāsa and viññāna - SN
18.19/II 248). We do not directly experience the qualities of re, water, air or earth.
We experience feelings resulting from the operations of our senses. To conceive our
body and the processes that happen there as the four or six dhātu-s requires a good
deal of deliberate, conceptual work. For example, as part of the contemplation of
the earth element, one should actively think and imagine that spleen is a particular
compound of the body, and being devoid of thought and rigid it must be considered
paṭhavī (Vism XI.64). It may normally not be directly experienced, probably apart
from the cases of medical conditions. Real-life vipassanā meditator Sunlun Shin
Vinaya seems to be well aware of that practical problem:
13 SN 35.1/IV 1: cakkhuṃ, bhikkhave, aniccaṃ. Yad aniccaṃ taṃ dukkhaṃ; yaṃ dukkhaṃ tad
anattā. Yad anattā taṃ ‘n’ etaṃ mama, n’ eso ’ham asmi, na m’ eso attā’ti evam etaṃ yathābhūtaṃ
sammappaññāya daṭṭhabbaṃ.
14 SN 35.25/IV 16: yam p’ idaṃ cakkhusamphassapaccayā uppajjati vedayitaṃ sukhaṃ vā
dukkhaṃ vā adukkhamasukhaṃ.

LIBERATING INSIGHT
97
They have to be approached through indirection, through the
repetition by word of mouth of the essential characteristics and
a forcing of understanding of their natures. This understanding
normally takes place rst in the realm of concepts.[…] If it were
true that it is necessary to handle the processes with the gloves of
concepts and thoughts, that processes can never be got at directly,
then there can be no path to freedom and no liberating knowledge
(Korneld, 1996: 90, 92).
Do not reect that this is rupa and this nama. Do not consider that
this is anicca, this dukkha and this anatta. All thinking, reection and
consideration are conceptual. They are not vipassana (Korneld,
1996: 104-105).
Of course this statement does not constitute a conclusive argument in itself,
but is nonetheless worth taking into consideration. There is also a second
very serious psychological problem connected with insight meditation which
is deliberately practised through the active applying of categories and active
analysis. While it attempts to provide a clear view of the mind-body complex as
it really is here and now, it fails to become aware that the very mental activity of
performing insight -- understood as: “deliberate analysis”, “applying categories”,
“maintaining active consciousness” -- is in itself an important mental process
which constitutes an essential part of our very being/selfhood in the moment.
Therefore these “mental” acts should probably themselves become the object
of insight in order for it to be complete and all-embracing. That is because such
mental states of performing active discursive insight would in fact be made up
(abhisankhata), and intended/planned (abhisañcetayita), as the Aṭṭhakanāgara
Sutta suggests about the jhānic states. But this would create a sort of “vicious
circle” of insight, and thus render the practice ineffective and in fact impossible.
Perhaps something else was however meant by the descriptions, such
as the above mentioned one from the Saḷāyatana Saṃyutta. Perhaps it is
simply a very specic way of saying that a meditator directly knows and sees
his body (kāya) as it really is. Then one of the several theoretical schemes
used to analyze the body into several components is applied, like that of
dhātu-s, āyatana-s, khandha-s or simply feelings (vedanā). This might not
necessarily mean, however, that if asked about the content of his experiences,
the meditator would reply using these terms; he might even not know them,

98
LIBERATING INSIGHT
and yet meditate correctly. In a similar way we could perhaps say that the
meditator directly knows and sees protons, neutrons and electrons constituting
his body as impermanent, painful and non-self. Isn’t his body constituted by
these elements? So if he really knows and sees his body as it really is, then
according to this particular mode of speaking, he must directly know them as
well. And yet they do not correspond to any elements of his direct experience
which could be verbalized for the sake of report. He may in all probability not
even know these concepts or even anything that directly corresponds to them.
So although the accounts which combine insight with jhāna give no indication
that they should not be taken rather literally, it could perhaps be argued
that these formulas were neither instructions for practice, nor immediate
verbalizations of direct experience. However, at some point they came to be
seen exactly as such and formed the scriptural basis for the so called “analytic
insight”, undoubtedly due to the growing confusion about the true nature of
Buddhist practice and of insight itself.
So we can see problems with concepts of insight knowledge and insight
practice in themselves and with their connection to jhāna meditation on
both the textual and the practical, psychological level. The psychological
implausibility connected with these concepts may be an important
hint suggesting that their very presence in the suttas is in fact a result
of doctrinal polemics, shifts and evolution and not of real life practice.
Alexander Wynne has suggested in The Origin of Buddhist Meditation that
at some point:
The scheme of jhānas became a support for different versions of
intellectual insight; meditation became the means for an increasingly
elaborate set of mental gymnastics (Wynne, 2007: 124).
A strictly philological approach may not take us much further. Professor
Johannes Bronkhorst has made a very important suggestion concerning this
problem, during a brief e-mail correspondence I once had with him. He has
suggested that even when by a purely philological approach we reach the
conclusion that certain accounts describe real meditational events, we still
have to make sense of them. This is a psychological problem, and we need a
psychological theory to provide proper interpretations of these accounts.
Therefore let us now turn for help to modern cognitive science. If the original
concept of liberating insight was indeed based on real life experience and thus

LIBERATING INSIGHT
99
on actual human cognitive processes, the results of this dynamically developing
discipline may shed some important light on our problem.
Liberating insight is supposed to be a special kind of understanding
bringing transformative knowledge. But what is the real cognitive mechanism
of insight? Where does it take place? In the stereotypical concept of Buddhist
insight we nd some implicit preconceptions which seem to be rooted in our
ordinary, common sense way of thinking. Let’s take for example this denition
by Gunaratana:
Insight meditation requires investigation and observation
(Gunaratana, 1985: 151).
Or by Grifths, from his paper: Concentration or Insight: The Problematic
of Theravāda Buddhist Meditation:
Wisdom is a type of […] discursive knowledge and vision.
The means used to achieve this kind of conscious awareness
is a continuous attempt to internalize the categories of Buddhist
metaphysics (Grifths, 1981: 613).
There is nothing particularly controversial or unique about these
denitions, as they represent a fairly stereotypical approach to insight,
quite popular in Buddhism. They are based on a common sense, widely
accepted notion of “understanding” as a deliberately undertaken mental
activity which takes place in a eld of consciousness. This understanding
results in obtaining declarative knowledge of explicit character – the
liberating knowledge which is so problematic in the suttas. This is common
sense. However, what modern cognitive science tells us, is that the way in
which we think, solve cognitive problems, come up with new ideas and
make decisions goes very much against what is considered to be common
sense. The widespread, common sense approach to these issues is often
labeled as “folk psychology”. According to folk psychology, higher level
cognitive operations are performed in a controlled, willed way on the basis
of consciousness. We have access to our higher cognitive operations, we
are aware of them, we actively make them happen through the effort of our
will. Although labeled “folk psychology”, this approach has in fact been a
prevalent trend also throughout the history of sophisticated Western thought,
for example in Cartesian philosophy.

100
LIBERATING INSIGHT
We can however see for ourselves the limitations of this model when it
comes to explaining how we arrive at some new ideas and yes, insights. These
are the well-known “eureka effects”, “a-ha moments” – when we suddenly
become aware of a new insight, without prior awareness of the process leading
to its emergence. This has led people in the past centuries to often attribute them
to some kind of divine inspiration. The explanation of this phenomenon and
many more of our mental operations became possible with new developments
in cognitive science fueled by research in neuroscience and in particular
the concepts of: unconscious information processing, implicit learning and
tacit knowledge. This unconscious is however, not the well-known Freudian
psychoanalytic unconscious - it is the cognitive unconscious, rst described by
Kihlstrom in 1987. James S. Uleman sums up this concept in his introduction to
the collective work entitled The New Unconscious:
In early models, the unconscious referred to pre-attentive
perceptual processes and latent memory traces, so that complex
higher mental processes depended on awareness for their operation.
In later models, complex processing did not require awareness
of the information that was transformed, so much more complex
unconscious cognitive processing occurs.
He (Kihlstrom) concluded that “conscious awareness…is not
necessary for complex psychological functioning” (Uleman, 2005:
5-6).
These new developments often seem to go strongly against the most basic
common sense notions. In one of the chapters of the same book, Ap Dijksterhuis,
Henk Aarts and Pamela K. Smith write:
First of all and strictly speaking, conscious thought does not
exist. Thought when dened as producing meaningful associative
consciousness, happens unconsciously. One may be aware of
some elements of the thought process or one may be aware of a
product of a thought process, but one is not aware of thought itself
(Dijksterhuis, Aarts, Smith, 2005: 81).
Our senses can handle about 11 million bits per second.[…]The
processing capacity of consciousness pales in comparison.[…] our
consciousness can process 50 bits per second.[…]In other words,

LIBERATING INSIGHT
101
consciousness can only deal with a very small percentage of all incoming
information. All the rest is processed without awareness. Let’s be grateful
that unconscious mechanisms help out whenever there is a real job to be
done, such as thinking (Dijksterhuis, Aarts, Smith 2005: 82).
This information may be of value when we consider the potential mechanism
of Buddhist insight. It was supposed to be concerned with the body (kāya), as
it really is (yathābhūtaṃ), connected with the unaltered awareness of the sense-
input. If our mental processing indeed works as the aforementioned cognitive
scientists would like us to believe, then this shows the limitations of conscious
awareness when it comes to interpreting sense data. Although Buddhist thought
seems to be actively against the notion of a conscious self, its stereotypical
concept of insight seems to rely on preconceptions of “folk psychology” which
are criticized by modern cognitive science.
Now I would like to draw attention to some remarks by Professor Ran R.
Hassin of the Hebrew University of Jerusalem. They are of particular value to
us, since they explicitly deal with the mechanism of cognitive insight.
It seems that the processes that yield insights do not require conscious
awareness. […]These ndings seem to suggest that insights tend
to pop up in awareness without prior conscious evidence for their
formation. […] In a series of studies we further examined whether
insights can occur not only in the absence of conscious awareness of
the processes that lead to them, but also in absence of the conscious
awareness of the insights themselves (Hassin, 2005: 204).
And indeed the occurrence of such insights was conrmed by the study.
Implicit insights are unconscious in that (a) they can occur without
awareness of the learned rules and (b) they can be manifested in
behavior without awareness. In addition, implicit insights are
unintentional in that they can occur when people do not intend
for them to happen, and in that they can affect behavior without a
corresponding intention (Hassin, 2005: 205).
I think it’s quite obvious that the notion of implicit insight presented here should
be of great interest to us, as it may provide a key to understanding how insight was
supposed to operate in the early Buddhist context. We have already seen that it is
the issue of the explicit content of liberating knowledge which is so problematic.

102
LIBERATING INSIGHT
Johannes Bronkhorst has stated that:
Prajñā referred to some unspecied and unspeciable kind of
insight (Bronkhorst, 1986: 102).
We must conclude, that if the earliest Buddhist tradition
acknowledged the existence of any liberating insight at all – and it
possibly did – this insight remained unspecied (Bronkhorst, 1986:
104).
Likewise, Wynne (2007: 123):
The content of liberating insight in the earliest teaching is unclear.
It therefore seems that the idea that implicit insights can occur without
awareness of the learned rules could indeed make sense in the context of early
Buddhism.
Even more important is that they can be manifested in behavior without
awareness. Buddhist soteriology is interested in achieving a fundamental
change in the human being, reaching to his very core. A change that results in
a completely new way of functioning in the world. A crucial element of this
new way of functioning is that it doesn’t require conscious, sustained effort to
be maintained – it happens spontaneously and effortlessly. Now the problem
with explicit, conscious learning which results in the attainment of verbal
knowledge, which in turn can later be declared, is that it usually doesn’t produce
permanent changes in our behavior, but on the contrary usually serves as a basis
for conscious, deliberate, long-term decision making. If the results of insights
can be manifested in behavior without awareness and if they can affect behavior
without a corresponding intention, that all ts well with the spontaneous,
effortless state of an arahant. The fact that insights are unintentional, in that
they can occur when people do not intend them to happen, would also solve
many interpretative problems, as there would be no need for performing any
additional, deliberate, consciously willed insight practice in the state of jhāna,
where such activities seem difcult to imagine.
Let us now briey summarize the results of our foray into the eld of cognitive
science, which to some may appear surprising or perhaps even unwarranted. As
we have seen, there are several very important problems with respect to the issue
of liberating insight and its connection to the meditative states of the four jhāna-s
in early Buddhism. Certainly this proved to be a huge interpretative problem for

LIBERATING INSIGHT
103
the later generations of Buddhists, and as some modern scholars have pointed out,
they attempted to solve this problem by introducing several concepts of theoretical,
deliberately practised insight which result in explicit knowledge, for example,
insight into the four noble truths. As we have seen however, applying such
concepts has resulted in psychological implausibility and in severe discrepancies
on the textual level. Schmithausen has stated that it seems reasonable to expect that
liberating insight was a psychologically plausible process (Schmithausen, 1981:
208). Taking into account all that we have learned about this elusive liberating
insight of early Buddhism, we should be looking for a psychological mechanism
which does not have to be deliberately and consciously practised, so that it could be
harmonized with a state such as the fourth jhāna, and which possesses unspecied
and perhaps unspeciable content (cf. Bronkhorst, 1986: 102, Wynne, 2007: 124).
It need not necessarily produce any explicit knowledge as a direct result, but should
still be cognitive in character and able to cause change in a human being. While
the theories of “implicit insight” or “cognitive unconscious” are not yet completely
developed in detail, and seem to have not been used so far with specic reference
to Buddhist meditation, they provide a very promising perspective: a possible hint
concerning the nature of the elusive liberating insight of early Buddhism.
Bronkhorst has recently made an interesting comment:
The relevant claims in the early Buddhist texts (which we will
consider in detail below) concern psychological states and processes
that are unusual from a commonsensical point of view. They are not,
however, in conict with any established rules of natural science
or psychology. […] The claims made in the early Buddhist texts
may not agree with the way we think about ourselves and other
human beings, but that may merely mean that we have to revise our
thoughts about ourselves (Bronkhorst, 2012: 73-74).
If liberating insight was indeed a cognitive mechanism similar to that
postulated by the modern cognitive scientists, then it would very much go
against common sense or “folk psychology”. This would make it easier for
us to understand how the later Buddhists could become confused regarding
such a supposedly essential element of their doctrine. This fact, coupled with
the external inuence of an environment which saw knowledge as liberating
(cf. Bronkhorst, 1986: 104, Vetter, 1988: XXXII-XXXIII), would then lead to
serious reinterpretations of the original doctrine.

104
LIBERATING INSIGHT
But could the four jhāna-s in their original form be endowed with at least
this kind of implicit insight? Wynne has suggested that the terms contained in
the stock description of the jhāna-s may be perhaps connected with insight/
awareness, and not with the rm keeping of an object in the mind, as traditional
interpretation would like to have it:
Words expressing the inculcation of awareness e.g. sati,
sampajāna, upekkhā are mistranslated or understood as
particular factors of the meditative states.[…] They give the
misleading impression that the third and fourth jhāna are
heightened states of meditative awareness characterized by
some sort of indescribable inner calm. But these terms have
quite distinct meanings in the early Buddhist texts: they
refer to a particular way of perceiving sense objects (Wynne,
2007:123).
This may not be in itself, however, a completely conclusive argument in our
case, as the term such as sati can be interpreted as pertaining to a perfect way
of concentrating on a meditative object, keeping it in mind. Such is in fact the
interpretation of Ajahn Brahm:
One’s mindfulness is greatly increased to a level of sharpness that
is truly incredible. One is immensely aware. Only mindfulness
doesn’t move. It is frozen (Brahm, 2006: 153).
The notion of intrinsic insight is not easily reconciled with the vision of
the four jhāna-s as a yogic type of meditation practised by concentrating on
single objects, leading to the shutting of the senses and to the stillness of the
mind. Fortunately, however, there is a good deal of argument that point to
the contrary.
In the Indriyabhāvanā Sutta (MN 152/III 298) we nd a critique of
meditative practice leading to the shutting down of the senses. The sutta
suggests as the apex of their development a state described by the same terms
as the third jhāna:
If he wishes that (sace ākaṅkhati) — “by having avoided
(abhinivajjetvā) both disagreeable/ objectionable (paṭikūla) and
agreeable (appaṭikūla) let me dwell (vihareyyaṃ) equanimous
(upekkhako) mindful (sato) and clearly comprehending

LIBERATING INSIGHT
105
(sampajāno),” he dwells there equanimous, mindful and clearly
comprehending.15
The terms upekkhako, sato, and sampajāno are exactly the same as the ones
used in the description of the third jhāna. The stock description of the fourth
jhāna speaks of giving up (pahāna) of pleasure (sukha) and pain (dukkha),
as well as of the earlier settling down (atthaṅgama) of somanassadomanassā
(mental ease and mental pain). This may very well correspond to avoiding
the agreeable and the disagreeable as described in the Indriyabhāvanā Sutta.
This would mean, just as Wynne would like to have it, that at least the third
and fourth jhāna were originally not meant to be states of sense-inactivity and
mental stasis.
The Mahātaṇhā-saṅkhaya Sutta (MN 38/I 256) suggests that after the
attainment of the fourth jhāna the meditator dwells with each of the six senses
registering its respective objects, mindfulness of the body established/present
(upaṭṭhitakāyasati), and with an immeasurable mind (appamāṇacetaso).
He dwells having abandoned both compliance and opposition
(anurodhavirodhavippahīno), and whatever feeling (vedanaṃ) he feels, he does
not delight in it (nābhinandati), does not welcome it (nābhivadati), and remains
not being bound to it (nājjhosāya tiṭṭhati).16 It seems that this sutta describes the
same state as the Indriyabhāvanā Sutta, just using different terms.
There is also a great deal of research by Bronkhorst (1986) showing the
original four jhāna-s as radically different from the mainstream methods. Vetter
suggests that the spontaneous jhāna of Early Buddhism became replaced with
a more articial one (Vetter, 1988: XXXVI). I have also attempted to provide
some additional evidence and reconstruct this process of reinterpretation of the
four jhānic states in my Reexamining Jhāna: Towards a Critical Reconstruction
of Early Buddhist Soteriology (Polak, 2011).
If that is the case, insight could be indeed an intrinsic quality of the four
jhāna-s, inherent in their very nature and method form their earliest phases, which
15 MN 152/III 301: sace ākaṅkhati — ‘paṭikūlañca appaṭikūlañca tadūbhayaṃ abhinivajjetvā
upekkhako vihareyyaṃ sato sampajāno’ti, upekkhako tattha viharati sato sampajāno.
16 MN 38/I 270: so cakkhunā rūpaṃ disvā piyarūpe rūpe na sārajjati, appiyarūpe rūpe na
byāpajjati, upaṭṭhitakāyasati ca viharati appamāṇacetaso. Tañ ca cetovimuttiṃ paññāvimuttiṃ
yathābhūtaṃ pajānāti — yatth’ assa te pāpakā akusalā dhammā aparisesā nirujjhanti. so evaṃ
anurodhavirodhavippahīno yaṃ kiñci vedanaṃ vedeti, sukhaṃ vā dukkhaṃ vā adukkhamasukhaṃ
vā, so taṃ vedanaṃ nābhinandati nābhivadati nājjhosāya tiṭṭhati.

106
LIBERATING INSIGHT
would increase its momentum with the progress of meditation, until reaching
its apex in the state of the fourth jhāna. According to the Kāyagatāsati Sutta
(MN 119/III 88) this apex would come about by having pervaded (pharitvā) the
body (kāya) of the meditator by means of a bright, puried mind (parisuddhena
cetasā pariyodātena).17 This activity would be spontaneous, not requiring any
deliberately undertaken conscious effort to understand or analyze it. Once the
citta had been puried, it would spontaneously effectuate insight, provided it
had the sense data to work with. Probably no separate conscious and willed
activity was needed or even possible in the state of the fourth jhāna.
The Cetanākaraṇīya Sutta (AN 10.2/V 2) states that for one who is
concentrated (samāhitassa) no intention ought to be made (na cetanāya
karaṇīyaṃ): "I know and see things as they really are" (‘yathābhūtaṃ jānāmi
passāmī’ti).18 It is a natural process that one should expect (dhammatā esā).
Then just as naturally come weariness/disenchantment (nibbidā), dispassion
(virāgo), cessation (nirodho) and liberation (vimutti).
When I presented the initial version of this paper at the IABS Conference
in Vienna in 2014, Alexander Wynne pointed out to me yet another sutta
which describes the spontaneous character of true insight: the Sekha Sutta
(MN 53/I 353). It describes the disciple of the noble ones (ariyasāvako) who is
endowed with virtue (sīlasampanno hoti), guards the doors of the sense faculties
(indriyesu guttadvāro hoti), observes measure in eating food (bhojane mattaññū
hoti), is devoted to wakefulness (jāgariyaṃ anuyutto hoti), attains the four
jhāna-s at will (nikāmalābhī), without difculty (akicchalābhī) and without
trouble (akasiralābhī). Such a person is compared to a hen (kukkuṭī) whose
eggs had been properly covered, warmed and incubated. There is no need for
any wish to arise (na evaṃ icchā uppajjeyya), in order for her chicks to break
17 MN 119/III 94: catutthaṃ jhānaṃ upasampajja viharati. so imam eva kāyaṃ parisuddhena
cetasā pariyodātena pharitvā nisinno hoti; nāssa kiñci sabbāvato kāyassa parisuddhena cetasā
pariyodātena apphuṭaṃ hoti.
18 AN 10.2/V 2: samāhitassa, bhikkhave, na cetanāya karaṇīyaṃ — ‘yathābhūtaṃ jānāmi
passāmī’ti. dhammatā esā, bhikkhave, yaṃ samāhito yathābhūtaṃ jānāti passati.

LIBERATING INSIGHT
107
through the eggshells and be born safely.19 Subsequently, having come to the
purity of equanimity and mindfulness (upekkhāsatipārisuddhiṃ āgamma), as a
result of the destruction of the efuents (āsavānaṃ khayā) he attains and abides
in efuent-free (anāsavaṃ) liberation of the mind (cetovimuttiṃ) and liberation
through understanding (paññāvimuttiṃ), which he has realized (sacchikatvā)
himself (sayaṃ) by direct knowledge (abhiññā). This is however likened to
a chick breaking out (abhinibbhidā) from an egg, so no volition or separate
“practice” is needed for that.20 And indeed, while all the preceding elements
of the Buddhist path ending with the four jhāna-s are described as a matter of
conduct (idam pi ’ssa hoti caraṇasmiṃ), the release from efuents is a matter of
knowledge (vijjā). This would mean that it cannot be “performed” – it “happens”
to oneself.
Such a type of insight would thus be implicit and unintentional, and not
produce any immediate verbal liberating knowledge which could be declared. It
would however produce profound changes in the cognition and functioning of
an awakened person. Its rst manifestation in declarable knowledge would be
the certitude of its profound effects – the knowledge of liberation. In this way
jhāna would prove to be a state with profound cognitive effects, thus justifying
its name, as jhāyati can be rendered as meaning “thinks”. It is however the
type of thinking suggested by the modern cognitive sciences that we are dealing
with here: occurring outside consciousness, having nothing to do with inner
speech (that would be vitakka), but nonetheless solving cognitive problems and
providing insights.
19 MN 53/I 356-357: yato kho, mahānāma, ariyasāvako evaṃ sīlasampanno hoti, evaṃ
indriyesu guttadvāro hoti, evaṃ bhojane mattaññū hoti, evaṃ jāgariyaṃ anuyutto hoti,
evaṃ sattahi saddhammehi samannāgato hoti, evaṃ catunnaṃ jhānānaṃ ābhicetasikānaṃ
diṭṭhadhammasukhavihārānaṃ nikāmalābhī hoti akicchalābhī akasiralābhī, ayaṃ vuccati,
mahānāma, ariyasāvako sekho pāṭipado apuccaṇḍatāya samāpanno, bhabbo abhinibbhidāya,
bhabbo sambodhāya, bhabbo anuttarassa yogakkhemassa adhigamāya. seyyathāpi, mahānāma,
kukkuṭiyā aṇḍāni aṭṭha vā dasa vā dvādasa vā tānāssu kukkuṭiyā sammā adhisayitāni sammā
pariseditāni sammā paribhāvitāni, kiñcāpi tassā kukkuṭiyā na evaṃ icchā uppajjeyya — ‘aho
vat’ ime kukkuṭapotakā pādanakhasikhāya vā mukhatuṇḍakena vā aṇḍakosaṃ padāletvā sotthinā
abhinibbhijjeyyun ’ti, atha kho bhabbā va te kukkuṭapotakā pādanakhasikhāya vā mukhatuṇḍakena
vā aṇḍakosaṃ padāletvā sotthinā abhinibbhijjituṃ.
20 MN 53/I 357-358: sa kho so, mahānāma, ariyasāvako imaṃ yeva anuttaraṃ
upekkhāsatipārisuddhiṃ āgamma āsavānaṃ khayā anāsavaṃ cetovimuttiṃ paññāvimuttiṃ diṭṭhe
va dhamme sayaṃ abhiññā sacchikatvā upasampajja viharati, ayam assa tatiyābhinibbhidā hoti
kukkuṭacchāpakasseva aṇḍakosamhā.

108
LIBERATING INSIGHT
We have arrived at a view of early Buddhist jhāna as a meditative practice
endowed with insight, maintaining the sensitivity of the mind and the senses,
and yet at the same time leading to altered states of consciousness free from
verbal, discursive thought. But is such a form of meditation possible at all?
Have we not come through our textual analysis to something nonsensical, an
oxymoron? It seems to be commonly accepted that there are two main types
of meditation: samatha, which leads to altered states of consciousness and to
stopping thought by concentrating on a single object, and vipassanā, which leads
to experiencing the world as it really is, but fails to bring a radically different
state of consciousness.
We may show that such a state is indeed possible, by pointing out that
somewhat similar forms of meditation can be tracked down in some later
Buddhist traditions and also in the teachings of some modern Theravādin
masters. One such meditative method is the practice of silent illumination
described in the teachings of the Chan master Hongzhi. Modern Chan Master
Sheng Yen recapitulates Hongzhi’s practice of silent illumination (mozhao) in
his book entitled Hoofprint of the Ox:
Hongzhi instructs his students to let go and settle quietly into
themselves, leaving behind all entangling conditions and supports
until they reach a point of perfect and unrestrained quiescence.
At the same time this does not imply that mind becomes dark or
incognizant. Quite the contrary, it is the distortions of deluded
and conditioned thinking that are silenced, not mental clarity or
awareness. With this silence, the mind’s innate wisdom shines
unobstructed, perfectly clear and luminous, without a single speck
of dust to impede it. “In this [state of] silent sitting”, Hongzhi says:
“the mind clearly perceives the details of sensory objects; yet, as
though transparent, no constructed image is produced” (Sheng-yen,
2001:142).
To begin with, silence and illumination are inseparable and must
be present simultaneously: in the very act of illumining, one
relinquishes grasping after thoughts and sensations, and directly
takes things in, thereby bringing the mind to perfect silence. […]
It is a mistake to think that rst one must develop inner calm, and,
only then, apply open awareness (Sheng-yen, 2001:147).

LIBERATING INSIGHT
109
Another meditative state which may be of interest to us is the vipassanā jhāna
described in the following way by the modern Theravādin master, Sayadaw U
Pandita:
Vipassanā jhāna allows the mind to move freely from object to
object, staying focused on the characteristics of impermanence,
suffering and absence of self, which are common to all objects.
Vipassanā jhāna also includes the mind that can stay focused
and xed upon the bliss of nibbāna. Rather than the tranquility
and absorption that are the goal of samatha jhāna practitioners,
the most important results of vipassanā are insight and wisdom.
[…] Vipassanā jhāna is the focusing of the mind on paramattha
dhammas. Actually they are just the things we can experience
directly through the six sense doors without conceptualization (U
Pandita, 1992:179).
I am not claiming here that the meditative states of silent illumination
and of the vipassanā jhāna are identical to early Buddhist jhāna. I am only
claiming that the very existence of such forms of meditation at least shows the
actual possibility of a state which can be simultaneously endowed with both
insight and calm, be devoid of verbal thoughts, and yet retain the sensitivity
of the body, without being attained by concentrative methods. This ts pretty
well with all the textual evidence we have about the four jhāna-s. The fact
that early Buddhist jhāna was seemingly such a paradoxical state must have
greatly contributed to its later fundamental misinterpretation. As Bronkhorst
(1986:88) has stated:
Already early in the history of Buddhism there was uncertainty about
the details of the practice taught by the Buddha. This uncertainty
opened the door to foreign elements which could take the place of
original but little understood elements.
The concept of implicit insight, which goes so strongly against common
sense, would seem to have been one of these little understood elements.
While the later Buddhists knew that insight was important, they could
only conceive it as producing a verbal form of knowledge, and different
doctrines came to be seen as direct expressions or verbalizations of such
knowledge. These doctrines in themselves were not inauthentic, but they

110
LIBERATING INSIGHT
were originally never supposed to be the expressions or direct verbalizations
of liberating insight. They probably originated as a result of deliberate,
conscious reflection upon the path leading to awakening or upon the new,
transformed way of cognizing and interacting with the world. But as
such, they could never perform any directly liberating function. Such was
probably the case with the doctrine of the four noble truths. But in the early
stages of the process, there was still awareness that the liberating insight
must be connected with the fourth jhāna. The presence of the knowledge of
encompassing the minds of others (cetopariyañāṇa) and of the knowledge
of recollecting past lives (pubbenivāsānussatiñāṇa) (both of which are
inauthentic in this context, as shown by Schmithausen (1981: 222)) after
the attainment of the fourth jhāna is probably also a result of attempts to
provide some accounts of declarative, explicit forms of knowledge which
could be vitally connected to liberation. A similar explanation has already
been proposed by Bronkhorst (1986: 115).
Subsequent to this, as the focus on contemplative insight intensified,
the four jhāna-s were reinterpreted, and no longer seen as a unique,
exclusively Buddhist practice, but as a concentrative, quasi-yogic
meditative method not very different from that of non-Buddhists. The
notion of liberating insight could not be easily harmonized with this new
vision of jhāna. One of the first attempts at providing this harmonization
was making the imperfections of the four jhāna-s the object of insight
practice. In the earlier phase of development the jhānic imperfections
were to be contemplated in the very state of jhāna, while later a rather
complicated concept of the vehicle of serenity was introduced. Probably
sensing the psychological difficulties of such a practice, the Buddhists
introduced the notion of purely theoretical insight, not connected in any
way to the four jhāna-s. As I have noted above, various formulas from
the Suttapiṭaka were then used as a basis for this new concept of insight.
And these formulas probably did not originally function as instructions
for practice, or immediate verbalizations corresponding to the elements of
our direct experience. This doctrinal evolution had resulted in a concept of
Buddhist meditation that could successfully function only on a theoretical
level. It is therefore no wonder that ultimately some Buddhists dispensed
with meditation altogether (cf. Wynne, 2007: 124, Polak, 2011: 174-190).

LIBERATING INSIGHT
111
Bibliography
Primary Sources:
References to Pāli texts are to the Pali Text Society editions.
AN Aṅguttara Nikaya
DN Dīgha Nikāya
MN Majjhima Nikāya
SN Saṃyutta Nikāya
Sn Suttanipāta
Vism Visuddhimagga
In this paper, the references to the Pāli suttas are to:
(before the slash)The number of the sutta/ (after the slash) the number of the PTS volume
and page in case of the suttas from the Dīgha Nikāya.
The number of the sutta/the number of the PTS volume and page in case of the suttas
from the Majjhima Nikāya.
The number of the saṃyutta and the number of the sutta/the number of the PTS volume
and page in case of the suttas from the Saṃyutta Nikāya.
The number of the verse in case of quotations from the Suttanipāta.
The number of the nipāta and the number of the sutta/the number of the PTS volume and
page in case of the suttas from the Aṅguttara Nikaya.
Secondary sources:
Bodhi, Bhikkhu. (2000) Abhidhammattha Sangaha. A Comprehensive Manual of
Abhidhamma. Pāli Text, Translation & Explanatory Guide, Onalaska: BPS
Pariyatti Editions.
Brahm, Ajahn. (2006) Mindfulness, Bliss, and Beyond: A Meditator’s Handbook,
Boston: Wisdom Publications.
Bronkhorst, J. (1986) The Two Traditions of Meditation in Ancient India, Stuttgart:
Steiner Verlag.
Bronkhorst, J. (2012) Absorption. Human Nature and Buddhist Liberation,
University Media.
Dijksterhuis A., Aarts H., Smith P. K. (2005) “The power of the subliminal: On

112
LIBERATING INSIGHT
Subliminal Persuasion and Other Potential Applications”, In Hassin R., Uleman
J. S., & Bargh J. W. (Eds.), The new unconscious (pp.77-106), New York:
Oxford University Press.
Grifths, P. (1981) “Concentration or Insight: The Problematic of Theravāda
Buddhist Meditation-Theory”, Journal of the American Academy of Religion
Vol. 49, No. 4 (Dec.), pp. 605-624.
Gunaratana, M. (1985) A Critical Analysis of the Jhānas in Theravāda Buddhist
Meditation, Washington: The American University.
Hassin R. (2005) “Non Conscious Control and Implicit Working Memory”, In
Hassin R., Uleman J. S., & Bargh J. W. (Eds.), The new unconscious (pp. 196-
223), New York: Oxford University Press.
Korneld, J. (1996) Living Dharma: Teachings of the Twelve Buddhist Masters,
Boston and London: Shambhala.
Polak, G. (2011) Reexamining Jhāna: Towards a Critical Reconstruction of Early
Buddhist Soteriology, Lublin: Wydawnictwo UMCS.
Rhys Davids, T. W., Stede, W. (2007), Pali-English Dictionary, Springeld: Nataraj
Books.
Schmithausen, L. (1981) “On Some Aspects of Descriptions or Theories of
‘Liberating Insight’ and ‘Enlightenment’ in Early Buddhism”, Studien zum
Jainism und Buddhism, eds. Klaus Bruhn and Albert Wezler, Wiesbaden: Franz
Steiner, pp. 199–250.
Sheng-Yen (2001) Hoofprint of the Ox: Principles of the Chan Buddhist Path as
Taught by a Modern Chinese Master, New York: Oxford University Press.
Uleman, J.S. (2005) “Introduction: Becoming Aware of the New Unconscious”. In
Hassin R., Uleman J. S., & Bargh J. W. (Eds.), The new unconscious (pp.3-15),
New York: Oxford Press.
U Pandita, Sayadaw (1992) In This Very Life: The Liberation teachings of the
Buddha: Wisdom Publications.
Vetter, T. (1988) The Ideas and Meditative Practices of Early Buddhism, Leiden: E.
J. Brill.
Wynne, A. (2007) The Origin of Buddhist Meditation, London and New York:
Routledge.