In this part, we will study the objections raised by the opponent against the definition of mithyAtva as beingsublated by GYAna.

advaita-siddhi text --------------------GYAnanivartyatvaM vA mithyAtvam.h |
nanu - uttaraGYAnanivartye pUrvaGYAne ativyAptiH,
mudgarapAtAdinivartye cha ghaTAdAv-avyAptiH,
GYAnatvena GYAnanivartyatva-vivaxAyAmapyayaM doShaH, adhiShThAna-
sAXAtkAratvena nivartye shuktirajatAdau cha GYAnatvena
GYAnanivartyatva-abhAvAt.h sAdhyavikalatA, GYAnatvavyApya-
dharmeNa GYAnanivartyatva-vivaxAyAM GYAnatvavyApyena smR^iti-
tvena GYAnanivartye saMskAre ativyAptiH -

Translation:

Definition: Alternatively, unreality is the property of being sublated by knowledge or cognition. Objection by oppponent: This definition of mithyAtva, ie. GYAnanivartyatvaM vA mithyAtvam, is too wide (ativyApti) in the case of a cognition's being sublated by a subsequent cognition. And in the case of a pot's being destroyed by stroke of a hammer, the definition would be too narrow (avyApti). If the intention is that unreality consists in being sublated by cognition as cognition, (GYAnatvena GYAnanivartyatva) even then there is the same defect, ie. of being too narrow. Further, in the case of the illusion of silver in nacre, the sublation of the illusory silver is due to the direct realization of the substratum, ie. the nacre, (adhiShThAna-sAxAtkAra), and there is no sublation due to cognition as cognition, which leads to the defect of sAdhyavikalatA, ie. the definition is defective. If the definition means that unreality is what is sublated by a property concomitant with the property of being a cognition (GYAnatva-vyApya-dharma), in the case of a memory (smR^iti) that sublates a mental impression (saMskAra), the definition is too wide.

As BrahmAnanda's GauDabrahmAnandI clarifies, this third definition is based on the shruti itself.

'vidvAnnAmarUpAdvimukta' ityAdishrutyarthe vivadamAnaM prati
sAdhyAntaramAha - GYAnanivartyatvaM veti |

A different sAdhya (thing to be proved) is (now) stated, as
per the interpretation of the shruti "The knower of Brahman is
free from names and forms.

Let us now examine the objection of the opponent point bypoint.

1) Definition is too wide-------------------------
First of all, he says that the definition has the defect of
being too wide (ativyApti). How? In the case of a cognition
being replaced by a subsequent cognition, the first cognition
though "out of focus" right now is certainly not false. Suppose,
I see the sun rising in the East and next I see a cow. At that
moment when I am cognizing the cow, the cognition of the sun is
in the background or out of focus, so to speak, being displaced
by the cognition of the cow. Does it mean my cognition of the sun
is false. No. The definition is too wide because it seems to apply
to this case where a cognition is displaced or sublated in some
sense by a subsequent cognition.

2) Definition is too narrow-----------------------------
Next, the opponent says that in the case of a pot being destroyed
by stroke of a hammer, thereby driving the pot into nonexistence,
the definition is too narrow. The advaitin claims that things such
as a pot are sublated by GYAna alone. Clearly, the opponent says,
the pot is not existent any more after being destroyed. So it should
be sublated. But what destroys or sublates the pot is not GYAna, but
something as simple as the stroke of a hammer! Actually, I
suspect that the dvaitin seems to be making fun of the advaitin
here - "You cannot make things such as a pot vanish by means
of what you call GYAna. If you want to make the pot vanish,
why don't you smash it with a hammer!"

Suppose a definition of a characteristic, say M, can be
expressed logically as M(X) <-> GN(X), which means X has
characteristic M if and only if X has characteristic GN.
Now, if we can find some Y such that Y has GN but not M,
then the definition is too wide and suffers from the defect
termed in nyAya as "ativyApti". In the case of a cognition C
which is sublated by another subsequent cognition, C', C has
the characteristic GN, GYAnanivartyatva, but not M, mithyAtva.
The definition here is too wide.

On the other hand, if we can find a Y such that Y has characteristic
M but not GN, then the definition is too narrow, a defect called
"avyApti". In the case of a pot, P, it is destroyed by a
stroke of a hammer and thereby has characteristic M, mithyAtva,
but not characteristic GN. It is not destroyed by any GYAna.
So the definition is too narrow in this case.

3) What if GYAnanivartyatva means GYAnatvena GYananivartyatva
--------------------------------------------------------------
The opponent now seems to anticipate a move by the advaitin
to silence the charge of ativyApti. Suppose the advaitin
says "By GYAnanivartyatva, we really mean that the cognition
(GYAna) that sublates what is mithyA has to be considered
as a cognition acting as a cognition, not anything else."

BrahmAnanda clarifies this "anticipated position" of
the advaitin in the language of navya-nyAya:

GYAnatvena GYananivartyatveti | GYAnatva-avachchhinna-
kAraNatApratiyogika-kAryatAvan-nAsha-pratiyogitvetyarthaH |

By "GYAnatvena GYAnanivartyatva" is meant the characteristic
of being the counterpositive (pratiyogin) of the destruction
which has an effect-ness, where the effect-ness is related
to a cause-ness delimited (determined) by cognition-ness.

The gist is that what determines the cause of destruction
of something unreal is the GYAna acting as a GYAna, ie. the
characteristic GYAnatva.

Now, suppose the anticipated position holds that cognition
of the nature of a vR^itti (a mental modification) is a quality
(dharma) of the mind (manas). The VedAnta-paribhAShA
says: "vR^itti-rUpa-GYAnasya manordharmatve cha" and cites
shruti as support for this position. Even in the nyAya system,
cognition is said to be a property (guNa) of the self.
(GYAnAdhikaraNam-AtmA - tarkasaMgraha of annaMbhaTTa). If cognition
is a quality of the mind (or self), any cognition that displaces a
previous cognition does so not in its capacity as a cognition but
in its capacity as a quality of the mind. So what happens
in this case is that one quality of the mind sublates another
quality of the self. Therefore, the definition of mithyAtva
is free from the defect of being too wide, simply because
the example of one cognition's being sublated by a subsequent
one is no longer relevant.

4) The charge of avyApti still holds-------------------------------------
The opponent now says that although the advaitin has freed his
definition from this defect of being too wide, the other defect
of being too narrow still applies. The GYAnatvena GYAnanivartya-
tva clarification still does nothing to remove the defect in
the case of the stroke of the hammer which destroys the pot.

5) The definition is flawed---------------------------
Further, there is another problem with the "GYAnatvena
GYAnanivartyatva" definition. In the case of the illusion
of silver in nacre (shukti-rajata), the cause of the destruction
of the illusion is not just any cognition of nacre, rather it
is the cognition of that nacre which is the substratum of the
illusory silver. While a person mistakes a rope for a snake
in the dark, it is not sufficient to end his illusion by
holding some other rope in front of him! What does end the
illusion is his perceiving the rope which he mistook for
a snake. In such cases of illusion, the destruction of the
illusion is not by a cognition acting in its capacity as a
cognition. However, the illusory thing is admitted to be
mithyA. So even though mithyAtva is present, GYAnanivartyatva
is absent in the illusory thing. This gives rise to the defect
of sAdhya-vikalatA, ie. the definition is defective because
we see that mithyAtva is really not defined as GYAnanivartyatva
but as "adhiShThAna-sAxAtkAratva", which makes the advaitin's
attempt to prove the GYAnanivartyatva version futile!

6) What if a sub-class of GYAna is intended
--------------------------------------------
The opponent now anticipates another move by the advaitin.
What if GYAnanivartyatva is taken to mean "GYAnatvavyApya-
dharmeNa GYAnanivartyatva", ie. what is illusory is sublated
by a sub-class of GYAna, not just GYAna in general. To understand
this definition, remember that GYanatva is a property. A property,
say P, that is concomitant with GYAnatva, say G, is a property
such that whenever P is present, G is also present. In other words,
P implies G. The property P is said to be GYAnatva-vyApya or
"pervaded by GYAnatva." So the anticipated position is that
what is illusory (mithyA) is sublated by a cognition of thetype P.

7) Still the definition is too wide----------------------------------------

There is still a problem with the definition, says the opponent.
Consider smR^iti or memory which is a sub-class of GYAna, ie.
a specific kind of GYAna. The tarkasaMgraha of annaMbhaTTa
defines smR^iti as "saMskAramAtrajanyaM GYAnaM smR^itiH", memory
or recollection is cognition caused by reminiscent impression
alone. So it is clear that recollections are a sub-class or
a specific kind of cognition. The opponent, however, contends
that whenever a recollection occurs, the corresponding
reminiscent impression (saMskAra) that caused it gets
destroyed. In this case, the saMskAra, though destroyed by the
smR^iti is not considered false. So, once again, we have a
case where the definition of mithyAtva is too wide.

This last objection by the opponent is quite shaky at the
outset. First of all, even the naiyAyikas are divided regarding
whether the saMskAra gets destroyed by the smR^iti that it gives
rise to. The older nyAya-vaisheShika school holds that the
saMskAra gets destroyed by the smR^iti but logicians of the
newer navya-nyAya school tend to side with the advaitins in
rejecting this theory. The VedAnta-paribhAShA says: na hi smR^iteH
saMskAranAshakatva niyamaH, smR^itidhArA-darshanAt.h -
there is no rule that recollection destroys saMskAra because
a flow of recollection is observed. Therefore, the very premise
of the objection is questionable.

But still, the opponent might persist in challenging the advaita
position that smR^iti does not destroy saMskAra. As we shall
see, MadhusUdana's reply makes the last objection irrelevant
even if this advaita position regarding smR^iti is notadmitted.

MadhusUdana begins his reply thus (as usual):

iti chenna |

If this is what you contend, we say no (your objections are not justified).