Nau caGy upadhAyA$ hrasvaH ONPANINI 74001
Shorten the nexttolast before Ni before caG.mmmmmmmmm ^ C+ 857

This special shortening makes Nijanta roots like nAyi kAri lAvi pAci get reduplicated into a heavy-light rythm in the luG --

kR + Nic + luG tip aT + kAri + caG + t''' a + kari + caG + t →caGi a + ka + kari + caG + t →sanyataH a + ki + kari + caG + t →NeraniTi a + ki + kar + at →kuhozcuH acikarat →dIrgholaghoH acIkarat "he made (him) do"

nI + Nic + luG tip →luGlaG aT + nI + Nic + tip → a + nI + Nic + t''' →ata::upa a + nai + Nic + t → a + nai + Nic + caG + t →ecoya a + nAy + Nic + a + t a + nay + Nic + a + t →NeraniTi a + nay + a + t →caGi ananayat →sanyataH aninayat →dIrgholaghoH anInayat "he made him lead"

Why did sanyataH work here? We are not before san.

Because rule sanvalla made the caG be san-like.

KAZIKA aGgasya iti vartate. caGpare Nau yad aGgam, tasya upadhAyA hrasvo bhavati. acIkarat. ajIharat. alIlavat. apIpavat. atra dvirvacanopadhA-hrasvatvayoH prAptayoH paratvAd upadhAhrasvatvam, tatra kRte dvirvacanam. iha tu mA bhavAnaTiTatiti nityatvAd dvitIyasya dvirvacanaM prApnoti, tathA sati hrasvabhAvino 'Ggasya akArasya upadhAtvaM vihitam iti hrasvo na syAt? naiSa doSaH. oNeH Rdit karaNaM jJApakaM nityam api dvirvacanam upadhA-hrasvatvena bAdhyate iti. Nau iti kim? caGy upadhAyA hrasvaH ity ucyamAne alIlavat ity atra vacana-sAmarthyAd antaraGgAm api vRddhim AdezaM ca bAdhitvA hrasvaH syAt. adIdapatityatra hrasvatvena puko bAdhaH syAt. apIpacad ity evam Adautu naiva syAt. caGi iti kim? kArayati. hArayati. upadhAyA iti kim? acakAGkSat. avavAJchat. tadedadupadhAgrahaNam uttarArtham avazyaM kartavyaM tad iha api hrasvatvaM nivartayati ityevam arthaM yena na avyavadhAnam ityetannAzrayitavyam iti. upadhA-hrasvatve NerNicyupasaGkhyAnam. vaditvantaM prayojitavAn avIvadat vINAM parivAdakena. yo 'sau Nau Nilopas tasya sthAnivadbhAvena aglopitvAt aGgasya hrasvo na prApnoti. NyAkRtinirdezAt siddham.

ur Rt ONPANINI 74007
(Optionally nexttolast) R RR to R (before Nic plus caG).mmmmmmmmm ^ C+ 858

When we choose not to apply this optional rule, nexttolast R RR will become Ir ar Ar by other rules --

With the nichclass root kRRt (which is listed in the mmw as kIrt) we ordinarily get R turned into Ir --

kRRt + luG tip kRRt + Nic + caG + t''' →upadhAyAzca kir + Nic + at →halica kIr + Nic + at acikIrtat "he said, praised"

and the R of vRt + causative Nic becomes ar --

vRt + Nic + at →puganta vart + Nic + at avavartat "he made it turn"

and the R of mRj becomes Ar --

mRj + Nic + at →mRjervRddhiH mArj + Nic + at amamArjat "he rubbed"

However, this rule says that we may choose to not apply those other rules, and when we don't, we replace R RR with R instead --

kRRt + luG tip kRRt + Nic + caG + t''' kRt + Nic + caG + t acIkRtat "he said, praised"

vRt + Nic + caG + t''' vRt + Nic + caG + t avIvRtat "he made it turn"

mRj + Nic + caG + t''' mRj + Nic + caG + t amImRjat "he rubbed"

Small details --

kRRt + luG tip kRt + Nic + caG + t →NeraniTi kRt + caG + t →caGi kR + kRt + at →urat ka + kRt + at →sanyataH ki + kRt + at →kuhozcuH ci + kRt + at →luGlaG acikRtat →dIrgholaghoH acIkRtat "he said, praised"

Notice that sanyataH worked here because sanvalla says that ka is san-like before kRtat, that has a light R.

KAZIKA Nau caGi upadhAyAH RvarNasya sthAne vA RkArAdezo bhavati. irarArAm apavAdaH. ir acikIrtat, acIkRtat. ar avavartat, avIvRtat. Ar amamArjat, amImRjat. vacanasAmarthyAdantaraGgA api irarAro bAdhyante. taprakaraNaM dIrghe 'pi sthAnini hrasva eva yathA syAt, acIkRtatiti. na ca ayaM bhAvyamAnaH, kintu AdezAntaranivRttyarthaM svarUpam eva etadabhyanujJAyate.

Rtazca saMyogAderguNaH ONPANINI 74010
R of cluster-starter to guNa (before liT).mmmmmmmmm ^ M- C+ 859

As in --

smR + Nal smar + a →ata::upa smAr + a →liTidhA sasmAra "he remembered"

smR + atus smar + atus →liTidhA sasmaratus "both remembered"

smR + ma smar + ma →kRsRbhRvR smar + ima →liTidhA sasmarima "we remembered"

Rule guNorti also makes R of clusterstarter into ar sometimes.

Why do we say "clusterstarter"?

This won't work on kR, that starts with k -- --

kR + atus →liTidhA cakR + atus →ikoyaNaci cakratus "both made"

kR + Nal →acoJNiti kAr + a →liTidhA cakAra "he made; I made"

Rcchaty RRtAm ONPANINI 74011
RR-enders, Rcch and R to ( guNa before liT).mmmmmmmmm ^ C+ 860

The Rcchati mentioned here is the root 06.0016 Rcch Rcha~ gatIndriyapralaya "go, go at, attack; faint".

Examples --

Rch + Nal →checa Rcch + a arcch + a →liTidhA a + arcch + a →ata::AdeH A + arcch + a →tasmAnnuDdvi An + arcch + a → Anarccha

Rch + atus Anarcchatus

Rch + atus Anarcchus

tRR + us tar + us →liTidhA tatarus →pecus terus

A zloka from the epics --

sAtyakiH kRtavarmANaGM kRtavarmA ca sAtyakim | AnarchatuzH zarair ghorais takSamANau parasparam || "And sAtyaki went at kRtavarman, and kRtavarman at sAtyaki, hacking each other with dreadful arrows."

But that one has archa not arccha.

As far as I know, rch and rcch are just different spellings of the same rcch sound. See my notes on checa.

KAZIKA Rcchater aGgasya, R ity etasya, RRkArAntAnAM ca liTi parato guNo na bhavati. Rccha Anarccha, AnarcchatuH, AnarcchuH. R AratuH, AruH. RRkArAntAnAm nicakaratuH, nicakaruH. nijagaratuH, nijagaruH. Rccher alaghUpadhatvAd aprApto guNo vidhIyate, RRtAM tu pratiSiddhaH. vRddhi-viSaye tu pUrvavipratiSedhena vRddhir eva iSyate. nicakAra. nijagAra.

zRR;dRR;prAM hrasvo vA ONPANINI 74012
zRR dRR pRR to short optionally (before a kit liT).mmmmmmmmm ^ C+ 861

Rule RcchatyRRtAm right above ordinarily replaces RR with ar before kit liT, as in --

pRR + atus →RcchatyRRtAm par + atus paparatus "both protected"

But these three roots may optionally replace with R instead --

pRR + atus pR + atus papratus "both protected"

Why does this rule bother to say "to short"? Had it been worded as just zRR;dRR;prAM vA, it would have made RcchatyRRtAm optional after these roots.

Good thinking. Yet, that wouldn't have worked, as we don't want RRta:: id dhAtoH and hali ca to kick in.

KAZIKA zRR dRR pRR ityeteSAM aGgAnAM liTi parato vA hrasvo bhavati. zRR vizazratuH, vizazruH. vizazaratuH, vizazaruH. dRR vidadratuH, vidadruH. vidadaratuH, vidadaruH. pRR nipapratuH, nipapruH. nipaparatuH, nipaparuH. hrasvavacanam itvotvanivRttyartham. kecidetat sUtraM pratyAcakSate. zrA pAke, drA kutsAyAM gatau, prA pUraNe ityeteSAm anekArthA dhAtavaH iti zRRdRRprAmarthe vartamAnAnAM vizazratuH, vizazruH, vidadratuH, vidadruH, nipapratuH, nipapruH ityetAni rUpANi sAdhayanti. tathA ca sati kvasau vizazRvAnityetad rUpaM na syAt.

R;dRzo 'Gi guNaH ONPANINI 74016
(R RR) -enders and dRz to guNa before aG aG'.mmmmmmmmm ^ C+ 862

Exception to kGitica.

Ordinarily, aG and aG', because of kGitica, don't cause guNa replacement --

sic' + luG tip asic + aG + t''' → asicat "he sprinkled"

bhid + aG' bhidA "a breaking"

This exception says that aG and aG' will still replace with guNa the R of dRz --

dRz + luG tip →iritovA dRz + aG + tip darz + aG + tip adarzat "saw"

and also all rootfinal R RR --

kR + luG tip akR + aG + t''' akarat "made, did"

jRR + aG' + Ap ( allowed by Sidbhid ) jar + a + A →akassa jarA- "old age"

KAZIKA RvarNAntAnAM dRzez ca aGi parato guNo bhavati. zakalAGguSThako 'karat. ahaM tebhyo 'karaM namaH. asarat. Arat. jarA. dRzeH adarzat, adarzatAm, adarzan.

asyates thuk ONPANINI 74017
asyati gets thuk (before aG).mmmmmmmmm ^ M- C+ 863

th(uk) is a kitaugment.

Example --

as' + luG tip → as + t''' →AD-aj-AdInAm A + as + t →akassa As + t →asyati;vakti As + aG + t Asth + aG + t → Asthat "he threw"

Similarly AsthatAm, Asthan.

KAZIKA asyater aGgasya thugAgamo bhavati aGi parataH. Asthat, AsthatAm, Asthan.

vaca:: um ONPANINI 74020
vac to voc (before aG).mmmmmmmmm ^ M- C+ 864 luG

The root vac gets aG before luG, by asyativaktikhyAtibhyoG.

This rule turns vac + aG into voc + aG.

Examples --

vac + luG tip →luGlaG avac + tip →asyativakti avac + aG + tip avoca + tip → avoca + t''' → avocat "he said"

vac + luG tas avoca + tAm''' → avocatAm "both said"

vac + luG jhi avoca + ant''' →saMyogAnta avocan "they said"

zIGaH sArvadhAtuke guNaH ONPANINI 74021
zIG to guNa before hard.mmmmmmmmm ^ C+ 865

All my examples here have kGit affixes, as before non- kGit, hardsoft would have made ze anyway.

Examples --

zIG + te''' zete "sleeps"

zIG + laT Ate''' ze + Ate →ecoya zayAte "both sleep"

zIG + laT ate''' →zIGoruT zIG + rate zerate "they sleep"

azIG + laG ata''' →zIGoruT azIG + rata azerata "they slept"

zIG + zAnac ze + Ana →ecoya zayAna- "that sleeps"

This rule won't work before liT, because liTca makes liT soft --

zI + liT ta zizI + ez →eranekA zizye "he slept"

zI + liT iT' zizI + e''' →eranekA zizye "I slept"

Why didn't aciznu work on zizye above?

Exception eranekAco.

ayaG yi kGiti ONPANINI 74022
zIG to zay before kGit y.mmmmmmmmm ^ C+ 866

This rule will work before soft only, because zIGassArvadhAtukeguNaH makes ze before all hards.

So before soft y kGit affixes, such as yak or lyap, we have --

zIG + bhAve laT ta → zI + yak + ta → zI + yak + te''' zayyate "sleep happens"

upa + zIG + lyap upazayya "after sleeping near"

a-kRt;sArvadhAtukayor dIrghaH ONPANINI 74025
Lengthen before yak, kyaG, yaG, soft yAsuT.mmmmmmmmm ^ C+ 867

Examples

before yak --

zru + before karmaNi → zru + yak zrUya

as in

zru + karmaNi laT ta → zrUya + ta → zrUya + te''' → zrUyate "is being heard"

ci + yak + laT jhi → cIyante "they are being piled up"

before soft yAsuT --

zru + @soft liG mip → zru + yAsam''' zrUyAsam "may I hear"

before yaG --

ci + yaG + laT ta ceci + ya + te''' cecIyate "piles up again and again, hoards"

before kyaG --

bhRzam + kyaG + laT ta →supodhA bhRza + ya + ta bhRzA + ya + ta → bhRzAya + te''' → bhRzAyate "becomes important"

But there is no lengthening before other y affixes, such as lyap or hard yAsuT --

pra + hR + ktvA → pra + hR + lyap →hrasvasyapitikRtituk pra + hRt + lyap → prahRtya "took away and"

su + @hard liG tip → su + znu + tip → sunu + yAt''' → sunuyAt "would squeeze"

KAZIKA akRdyakAre asArvadhAtukayakAre ca kGiti parato 'jantasya aGgasya dIrgho bhavati. bhRzAyate. sukhAyate. duHkhAyate. cIyate. cecIyate. stUyate. toSTUyate. cIyAt. stUyAt. akRtiti kim? prakRtya. prahRtya. paratvAt dIrghatvena tuko bAdhaH syAt. asArvadhAtuke iti kim? cinuyAt. sunuyAt. kGiti ityeva, uruyA. dhRSNuyA.

What's the literal translation of this rule?

Including trickle and rewording a bit, it says -- "lengthen before non- kRt soft kGit affixes that start with y". Methinks, the only such affixes are the four listed above.

What about the yaGluk?

It doesn't start with y. Rule sthAnivad Adezo 'n-al-vidhau says so.

In your bhRzAyate example above, shouldn't kartarizap have worked?

Oopsie sorry. I forgot to mention that. In fact, it did work. I should have added this --

bhRzAya + te''' →kartarizap bhRzAya + zap + te →atoguNe bhRzAyate

cvau ca ONPANINI 74026
( lengthen) before cvi too.mmmmmmmmm ^ C+ 868

Examples --

zucy azuci karoti →abhUtatadbhAve zuci + cvi + karoti zucI + v + karoti →verapRktasya zucIkaroti "cleans"

paTv apaTu bhavati →abhUtatadbhAve paTu + cvi + bhavati paTU + v + bhavati →verapRktasya paTUbhavati "becomes sharp"

See also asyacvau.

KAZIKA cvi-pratyaye parato 'jantasya aGgasya dIrgho bhavati. zucIkaroti. zucIbhavati. zucIsyAt. paTUkaroti. paTUbhavati. paTUsyAt.

riG za;yag;liGkSu ONPANINI 74028
R to ri(G) before za, yak, ( flat soft) liG.mmmmmmmmm ^ C+ 869

Roots made of one consonant plus R turn their R into ri before certain affixes.

Before za, this is an exception to ikoyaNaci --

mR + kartari te''' → mR + za + te mri + a + te →aciznu mriyate "dies"

yak --

mR + bhAve te''' → mR + yak + te mriyate "there is death here, someone is dying"

kR + yak + te''' kri + ya + ta kriyate "it is being done"

liG --

kR + soft liG tip → kR + yAt''' kri + yAt''' → kriyAt "may he do"

This won't work when R follows cluster -- see guNortisa below.

KAZIKA RkArAntasya aGgasya za yak ityetayoH liGi ca yakArAdau asArvadhAtuke parato riG ityayam Adezo bhavati. za - Adriyate. Adhriyate. yak - kriyate. hriyate. liG - kriyAt. hriyAt. riGvacanaM dIrghanivRttyartham. asArvadhAtuke ityeva - bibhRyAt. yi ityeva - kRSISTa. hRSISTa.

What about other R-enders?

See next rule.

Why is the replacement called ri(G), with G label?

So that it does not replace the whole root. See Gicca.

Why do you translate " flat soft liG" when the original just says liG? This rule gets " soft" from the trickle, but not " flat".

We are getting the trickle word yi "before y" from ayaGyikGiti. This trickle is irrelevant with za, that never starts with y, and with yak, which always does. But a soft liG replacement only starts with y if it has yAsuT, that is to say, if it is flat. So, this rule won't affect those that got sIyuT.

guNo 'rti;saMyog'-AdyoH ONPANINI 74029
Replace R of arti and cluster-starters with ar (before yak and flat soft liG).mmmmmmmmm ^ C+ 870

Previous rule replaces the R of roots made of one consonant plus R with ri. The other R ender roots replace R with ar.

Examples with arti --

R + yak + te''' aryate "it is reached"

R + yAt''' aryAt "may he reach"

Examples with smR, a cluster-plus-R root --

smR + yak + te''' smaryate "it is remembered"

smR + yAt''' smaryAt "may he remember"

Previous rule says za yak liG, but you only translated yak liG here. Shouldn't za be trickling too?

Actually it is. But that isn't worth translating, as none of the zaclass roots is made of a cluster plus R.

asya cvau ONPANINI 74032
A a (to I) before cvimmmmmmmmm ^ C+ 871

Examples --

bheka + cvi + karoti bhekI + v + karoti →vera bhekIkaroti "turns (him) into a frog"

khaTvA + cvi + karoti khaTvI + v + karoti →vera khaTvIkaroti "makes (it) into a bed"

See also cvauca.

KAZIKA I iti vartate. acarNAntasya aGgasya cvau parataH IkArAdezo bhavati. zuklIbhavati. zuklIsyAt. khaTvIkaroti. khaTvIsyAt.

kyaci ca ONPANINI 74033
(A a to I) before kyac.mmmmmmmmm ^ C+ 872

Examples --

putram + kyac →supodhA putra + kyac putrIya "want to have a son"

yAnam + kyac + laT tip yAnIyati "wants to have her own car"

KAZIKA asya iti vartate. kyaci parataH avarNAntasya aGgasya IkArAdezo bhavati. putrIyati. ghaTIyati. khaTvIyati. mAlIyati. akRtsArvadhAtukayor dIrghaH 7-4-24 ityasya apavAdaH. pRthagyogakaraNam uttarArtham.

dyati;syati;mA;sthAm it ti kiti ONPANINI 74040
do so mA sthA to i before kit t.mmmmmmmmm ^ C+ 873

Examples --

ava + so + kta ava + si + ta → avasita- "one that has settled"

mA + kta mita- "measured"

sthA + kta sthita- "staying"

sthA + ktavatu sthitavat- "stayed"

sthA + ktvA sthitvA "stayed and"

KAZIKA dyati syAti mA sthA ity eteSAm aGgAnAm ikAr'-Adezo bhavati takArAdau kiti pratyaye parataH. dyati tirditaH. nirditavAn. syati avasitaH. avasitavAn. mA mitaH. mitavAn. sthA sthitaH. sthitavAn. ti iti kim? avadAya. kiti iti kim? avadAtA.

dadhAter hiH ONPANINI 74042
dhA to hi (before kit t)mmmmmmmmm ^ C+ 874

Only examples --

dhA + kta → hi + kta → hita- "was put"

hitavat- "he put"

hitvA "put and"

KAZIKA dadhAter aGgasya hi ity ayam Adezo bhavati takAr´Adau kiti pratyaye parataH. hitaH. hitavAn. hitvA.

do dad ghoH ONPANINI 74046
dA of ghu to dath (before kit t).mmmmmmmmm ^ C+ 875

Replace the dA of a ghu with dath- before a kit affix that starts with t. As in --

dA + kta dath + ta →kharica datta- "given"

dA + ktvA dath + tvA →kharica dattvA "after giving"

Why do we say ghu roots?

The dA(p) that means "cut" is unaffected --

dAtamM barhiH "the barhis grass was reaped"

How do you know that the second nounbase in the original rule is dath-, and not dad- dat- or dadh-?

The kAzikA clears away that doubt. Read it carefully.

KAZIKA dA ityetasya ghu-saMjJakasya dad ity ayam Adezo bhavati takArAdau kiti pratyaya parataH. dattaH. dattavAn. dattiH. daH iti kim? dhItaH. dhItavAn. dheTaH etad rUpam. ghoH iti kim? dAp lavane dAtaM barhiH. daip zodhane avadAtam mukham. ayam AdezaH thAntaH iSyate. evaM hy uktam tAnte doSo dIrghatvaM syAd dAnte doSo niSThAnatvam. dhAnte doSo dhatva-prApti-sthAnte 'doSas tasmAt thAntam. yadi tu dasti iti takArAdau dIrghatvaM tadA tAnte 'pi adoSaH. dAntadhAntayor api sannipAtalakSaNo vidhiranimittaM tad-vighAtasya iti natva-dhatve na bhaviSyataH iti na doSaH. avadattaM vidattam ca pradattaM cAdikarmaNi. sudattam anudattaM ca nidattam iti ceSyate. aca upasargAt taH iti prApte nipAtyante. anupasargA vA ete avAdayaH kriyAntara-viSayA veditavyAH.

aca:: upasargAt taH ONPANINI 74047
After a vowel preverb, (replace dA of a ghu) with t (before kit t).mmmmmmmmm ^ C+ 876

Examples --

pra + dA + kta pra + t + kta → pratta- "given"

AG + dA + kta →doda Adatta- Atta- "taken"

This won't work after non- preverbs --

dadhi dattam "curds were given"

(Where datta- is dA + kta, with dodadghoH)

And won't work after a preverb that ends in a consonant --

/nis + datta- → nirdatta- "given out"

And won't work after daip, that has dA but is not a ghu --

avadAttam mukhyam

KAZIKA ajantAdupasargAduttarasya dA ityetasya ghusaMjJakasya ta ityayamAdezo bhavati takArAdau kiti. prattam. avattam. nIttam. parIttam. acaH iti kim? nirdattam. durdattam. upasargAtiti kim? dadhi dattam. madhu dattam. ghoH ityeva, avadAtam mukham. upasargAtiti paccamInirdezAdAderalaH prApnoti? tatra samAdhimAhuH. acaH ityetad dvirAvartayitavyam, tatra ekaM paccamyantaM upasargavizeSaNArtham, aparamapi SaSThyantaM sthAninirdezartham ityAkArasya sthAne takAro bhavati. dvitakAro vA saMyogo'yamAdizyate, so'nekAltvAt sarvasya bhaviSyati. apo bhi 74048 ityatra paJcamyantamacaH ityanuvartate. tena pakAramAtrasya bhaviSyati. dyaterittvAdacasta ityetad bhavati vipratiSedhena. avattam. prattaM juhoti.

apo bhi ONPANINI 74048
ap- to (d) before bh.mmmmmmmmm ^ C+ 877

Only examples --

ap- + bhis adbhis "with waters"

ap- + bhis adbhyas "from waters"

KAZIKA ab ity etasyAGgasya bhakArAdau pratyaye parataH ta ityayam Adezo bhavati. adbhiH. adbhyaH. bhi iti kim? apsu. svavaHsvatavasormAsa uSasazca takArAdeza iSyate chandasi bhakArAdau. svavadbhiH. svatavadbhiH. mAdbhiriSTvA indro vRtrahA. samuSadbhirajAyathAH.

Shouldn't this work before bhyAm too?

Yes, but ap- is only used before plural sup.

You mistranslated. Doesn't the original rule tell us to replace with t not with d?

It does. But jhalAMjazonte always turns that t into d.

saH sy ArdhadhAtuke ONPANINI 74049
s to ( t ) before soft s.mmmmmmmmm ^ M- C+ 878

Example before sya --

vas + lRT tip → vas + sya + ti vatsyati "will reside"

before sic --

vas + luG tip →luGlaG avas + tip → avas + sic + tip → avas + sic + t''' →vadavraja avAs + sic + t → avAs + s + It''' avAtsIt "he resided"

before san --

ghas + san ghat + sa →sanyaGoH ghaghatsa →sanyataH ghighatsa →kuhozcuH jhighatsa →abhyAsecarca jighatsa "want to eat"

What happens before hard s?

Nothing --

As + laT thAs → As + se''' → Asse "you are sitting"

tAs;astyor lopaH ONPANINI 74050
chop tAs and as (before s).mmmmmmmmm ^ M- C+ 879

Example with tAs --

kR + luT sip → kR + tAs + si →hardsoft kartAs + si kartAsi "you'll make"

Example with as --

as + laT sip asi "you are"

Why doesn't the word "before soft" trickle here from previous sutra?

That would make the rule useless, because no s-starter soft affix ever follows tAs or as. Common sense stops trickles.

ri ca ONPANINI 74051
chop tAs before r-affixes.mmmmmmmmm ^ M+ C+ 880

Only four examples --

tAs + tas →luTaHpratha tAs + rau tArau

tAs + AtAm → tAs + rau tArau

tAs + jhi →luTaHpratha tAs + ras tAras

tAs + jha → tAs + ras tAras

Some verbs made with these --

kR + luT tas → kR + tArau →hardsoft kartArau "both'll make"

kR + luT jhi → kR + tAras →hardsoft kartAras "they'll make"

ha:: eti ONPANINI 74052
( tAs and as to) h before e'''.mmmmmmmmm ^ M- C+ 881

As in

kR + luT iT' →syatAsI kR + tAs + i →hardsoft kartAs + i → kartAs + e''' kartAhe "i shall make"

as + laT iT' → ase →znasorallopaH se he

A verb "he"? Never heard of it. When does asti get bent endings?

As far as I know that only happens in vi + ati + as, allowed by kartarikarmavyatihAre. As in --

vyatihe "I do the duty of someone else"

sani mI;mA;ghu; rabha;labha;zaka; pata;padAm aca is ONPANINI 74054
Before ( iT-less) san, vowel of mI;mA; ghu; rabh;labh;zak; pat;pad to is.mmmmmmmmm ^ C+ 882

This changes the root vowel into is. In mI mA ghu, the s is before san, so sassyA replaces it with t --

mI + san mis + sa →sassyA mitsa "want to traverse"

dA + san dis + sa →sassyA ditsa "want to give"

In the other roots, the s is before a cluster, so skossa deletes it --

zak + san zisk + sa →skossa ziksa →kric zikSa "want to be able"

There is no reduplication, because of atralopo.

So we say --

ditsati "wants to give"

Alipsate "wants to get"

zikSati "wants to be able or competent"

pitsati "wants to fly"

KAZIKA sani pratyaye sakArAdau parataH mI mA ghu rabha labha zaka pata pada ityeteSAm aGgAnAm acaH sthAne isityayam Adezo bhavati. mI iti mInAtiminotyoH dvayor api grahaNam iSyate. mitsati. pramitsati. mA iti gAmAdAgrahaNeSu avizeSaH. mitsate. apamitsate. ghu ditsati. dhitsati. rabha Aripsate. labha Alipsate. zaka zikSati. pat pitsati. pada prapitsate. sani iti kim? dAsyati. si ityeva, pipatiSati. tani-pati.daridrANAm upasaGkhyAnam iti pateH iD-Agama-vikalpaH. sani rAdho hiMsAyAm aca is vaktavyaH. pratiritsati. hiMsAyAm iti kim? ArirAtsati.

Why didn't sanyaGoH add a stammer, as usual before san?

It did, but exception atra lopo 'bhyAsasya below ate it.

Ap;jJapy;RdhAm It ONPANINI 74055
(Vowel) of Ap jJapi Rdh to I.mmmmmmmmm ^ C+ 883

Exception to previous rule sanimImA, that would have replaced with is.

All examples --

Ap + san + laT tip Ipsati "wants to get (to)"

jJapi jJIpsati "wants to make (him) know"

Rdh Irtsati "wants to succeed"

There is no stammer, because of atra lopo 'bhyAsasya.

KAZIKA Ap jJapi Rdha ity eteSAm aGgAnAm aca IkArAdezo bhavati sani sakArAdau parataH. Ab Ipsati. jJapi jJIpsati. RdH Irtsati. jJapeH dvAvacau, tatra NeH pUrvavipratiSedhena lopaH, itarasya tu Itvam. sani ityeva, prApsyati. si ityeva, jijJapayiSati. ardidhiSati. sanIvantardha iti jJapeH Rdhezca iTo vikaopaH.

dambha:: ic ca ONPANINI 74056
(Vowel) of dambh to i toommmmmmmmm ^ C+ 884

The vowel of dambh "hurt, deceive" turns into I or i before iT-less san.

Only examples --

dambh + san dimbh + sa →aniditA dibh + sa →throwback dhibh + sa →jhalAMjazjhazi dhibsa →sanyaGoH didhibsa →atralopo dhibsa →kharica dhipsa "want to hurt"

dambh + san dImbh + sa dhIpsa "want to hurt" ( kharica (same steps)

The stammer disappeared by atra lopo 'bhyAsasya.

Because of sanIvantardhabhrasjadambhuzrisvRyUrNubharajJapisanAm, san becomes iSan optionally after dambh. In that case this rule does not work --

dambh + san → dambh + iSan →sanyaGoH dadambhiSa →sanyataH didambhiSa "want to hurt"

Here an-iditA did not work because iSan is not kGit (see halantAcca).

KAZIKA dambheH aca ikArAdezo bhavati, cakArAtIt ca sani sakArAdau parataH. dhipsati, dhIpsati. si ityeva, didambhiSati.

If aniditA works on dimbh, then san must be a kGit here. What made it kGit?

halantAcca did.

muco 'karmakasya guNo vA ONPANINI 74057
(Vowel) of objectless muc to guNa optionally.mmmmmmmmm ^ C+ 885

san is kit after muc (by halantAcca), so ordinarily we'd have no puganta. This exception however makes guNa anyway. So we may say --

muc + san + laT ta mokSate "he wants to free himself"

There is no stammer, because of atra lopo 'bhyAsasya.

If we don't take this option --

muc + san + laT ta mumukSate "he wants to free himself"

KAZIKA muco 'karmakasya guNo vA bhavati sani sakArAdau parataH. halantAc ceti kittva-pratiSedho vikalpyate. mokSate vatsaH svayam eva, mumukSate vatsaH svayam eva. akarmakasya iti kim? mumukSati vatsaM devadattaH. karmakartari mucir akarmako bhavati, karma-vizeSasya avivakSitAd vA.

atra lopo 'bhyAsasya ONPANINI 74058
And delete the stammer.mmmmmmmmm ^ C+ 886

The previous four rules, namely --

sanimImA

ApjJa

dambha_icca

mucokarma

delete the stammer. See examples under each rule.

KAZIKA yadetat prakrAntaM sani mImA ityAdi muco 'karmakasya iti yAvat, atra abhyAsalopo bhavati. tathaiva udAhRtam. abhyAsasya ityetac ca achikRtaM veditavyam A adhyAyaparisamApteH. ita uttaraM yad vakSyAmaH abhyAsasya ityevaM tad veditavyam. vakSyati, hrasvaH DuDhaukiSate. tutraukiSate. sani mImAdhurabhalabhazakapatapadAmaca is, abhyAsalopazca, ityevam siddhe yadatragrahanam iha akriyate, tad viSayAvadhAraNArtham, atraiva abhyAsalopo bhavati, sanvadbhAvaviSaye na bhavati. amImapat. adIdapat. sanvallaghuni caGpare 'naglope iti sanvadbhAvAt prApnoti. sarvasya abhyAsasya ayaM lopaH iSyate, tadartham eva kecitatragrahaNaM varNayanti. nAnarthake 'lo 'ntyavidhiH ityapare sarvasya kurvanti.

hrasvaH ONPANINI 74059
Shorten ( stammer).mmmmmmmmm ^ M+ C+ 887

Examples --

pU + us →liTidhA pU + pU + us pupU + us →aciznu pupuvus "they cleaned"

jIv + Nal →liTidhA jI + jIv + Nal jijIva "was alive"

Because of eca::igghra, shortening an ec makes i u --

nI + liT tip → nI + Nal →acoJNiti nai + Nal →liTidhA nai + nai + a ni + nai + a →ecoya ni + nAy + Nal → ninAya "he led"

iS + liT tip →puganta eS + Nal →liTidhA e + eS + Nal i + eS + Nal →abhyAsasyA iy + eS + a → iyeSa "he wanted"

kSip + Nal → kSe + kSep + Nal →hrasvaH kSi + kSep →halAdiHzeSaH ki + kSep + Nal →kuhozcuH cikSepa "he threw"

KAZIKA hrasvo bhavati abhyAsasya. duDhaukiSate. tutraukiSate. DuDhauke. tutrauke. aDuDhaukat. atutraukat. abhyAsasya anaci. abhyAsasya yaducyate anaci tad bhavati iti vaktavyam. carAcaraH. calAcalaH. patApataH. vadAvadaH. halAdiH zeSo na bhavati.

hal AdiH zeSaH ONPANINI 74060
consonant at start (of stammer) stays.mmmmmmmmm ^ M- C+ 888

In other words: " trash all the consonants of the stammer, except the one that is at the beginning, if any. "

Examples --

kSip + us →liTidhA kSi + kSip + us ki + kSip + us →kuhozcuH cikSipus "they threw"

kR + kR →urat ka + kR →uraNra kar + kR ka + kR →kuhozcuH cakR

smR + Nal →acoJNiti smAr + a →liTidhA smA + smAr + a →hrasvaH sma + smAra sasmAra "he remembered"

smR + us →Rtazca smar + us →liTidhA sma + smar + us sasmarus "they remembered"

See exception zarpUrvAH below.

KAZIKA abhyAsasya halAdiH ziSyate, anAdir lupyate. jaglau. mamlau. papAca. papATha. ATa, ATatuH, ATuH. AdizeSanimitto 'yam anAder lopo vidhIyate, tatra abhyAsajAteH AzrayaNAt kvacidapi vartamAno halAdiH anadeH sarvatra nivRttiM karoti. apare tu bruvate, zeSazabdo 'yaM nivRttyA viziSTam avasthAnam Aha. tadavasthAnam uktito yadyapi pradhAnam, avidheyatvAt tu tadapradhAnam. nivRttireva tu vidheyatvAt pradhAnam. tatra ayam artho 'sya jAyate, abhyAsasya anAder halo nivRttiH bhavati iti. sA kim iti AderavidheyAM satIm anivRttim apekSiSyate iti.

Why did hardsoft turn smR into smar before us? That us is kit, by asaMyogAlliTkit, so kGitica should have stopped hardsoft.

hardsoft did not work there. Click the arrow between smR + us and smar + us, okay?

zar-pUrvAH khayaH ONPANINI 74061
But the stop after zar (stays).mmmmmmmmm ^ M- C+ 889

Exception to halAdizzeSaH --

" If a stammer starts with zar plus stop, only the stop stays, and the other consonants are removed. "

For instance, the stammer spar keeps the p and loses the s and the r --

spRz + us →liTidhA spR + spRz + us →urat spa + spRz + us →uraNra spar + spRz + us pa + sparz + us → pasparzus "they touched"

And sthA keeps the th and loses the s --

sthA + us →liTidhA sthA + sthA + us thA + sthA + us →hrasvaH thasthA + us →abhyAsecarca tasthA + us →usya tasthus "they stayed"

This rule applies only to roots that startwith zar plus stop. The roots with a stop before a zar reduplicate normally --

kSip + us cikSipus "they threw"

and so do the ones that start with zar plus non- stop --

smR → sasmAra "he remembered"

KAZIKA abhyAsasya zarpUrvAH khayaH ziSyante, anye halo lupyante. cuzcyotiSati. tiSThAsati. pispandiSate. zarpUrvAH iti kim? papAca. khayaH iti kim? sasnau. kharpUrvAH khaya iti vaktavyam. ucicchiSati ity atra uccheH antaraGgatvAt tuki kRte dvirvacanam, tatra halAdiH zeSe sati abhyAse takAraH zrUyeta.

ku;hoz cuH ONPANINI 74062
ku and h (of stammer) to cu.mmmmmmmmm ^ M- C+ 890

When a stammer has k kh g gh h we replace it with the most-alike cu, namely with c ch j jh jh. The ch and the jh will become c j by other rules.

Example with a k root --

kSi ( before liT ) →liTidhA kSi + kSip →halAdizze ki + kSip cikSip

As in --

kSip + us → cikSip + us → cikSipus "they threw"

kSip + Nal →puganta kSep + Nal →liTidhA kSe + kSep + a →halAdizze kekSep + a →hrasvaH kikSepa cikSepa "he threw"

With a h root --

hu + hu ju + hu → juhu

As in --

hu + laT tip → juhu + ti →hardsoft juhoti "he pours"

juhvati "they pour"

KAZIKA abhyAsasya kavargahakArayoH cavargAdezo bhavati. cakAra. cakhAna. jagAma. jaghAna. hakArasya jahAra. jihIrSati. jahau.

The rule teaches "replace h with cu", therefore rule most-alike says, we must take the one of c ch j jh J that is most h-like. Is j the most alike?

No. If you are pronouncing your h c ch j jh correctly, the most-alike to h is jh. So this rule turns h into jh. Then rule abhyAsecarca replaces that jh with j.

ur at ONPANINI 74066
R RR L (of stammer) to a.mmmmmmmmm ^ M- C+ 891

Examples --

kR + liT mas → kR + ma →liTidhA kR + kR + ma ka + kR + ma →kuhozcuH cakRma "we did"

kLp + liT ta → kLp + ez →liTidhA kL + kLp + e kakLpe →kuhozcuH cakLpe "it fitted"

(See also exceptions sanyataH sanvalla artipipartyozca.)

Wait. You mistranslated. While discussing rule uraNra, you said that this rule must be translated as "R RR L to ar al".

Do'h. Yer rite. I stand corrected. Yes, uraNra worked. Above, I should have written this --

kR + liT mas → kR + ma →liTidhA kR + kR + ma kar + kR + ma →halAdizzeSaH kakRma →kuhozcuH cakRma "we did"

dIrgha:: iNaH kiti ONPANINI 74069
Lengthen (the stammer) of iN' before kit liT.mmmmmmmmm ^ C+ 892

Examples before atus us, that are kit by asaMyogAlliTkit --

iN' + atus →iNoyaN y + atus →liTidhA i + y + atus Iyatus "both went"

iN' + us Iyus "they went" (same steps)

Counterexamples before Nal thal, that are akit --

iN' + Nal →acoJNiti ai + a →liTidhA ai + ai + a →hrasvaH i + ai + a →ecoya i + Ay + a →abhyAsasyA iyAya "he went"

iN' + thal → i + ithal →hardsoft e + itha →liTidhA e + e + itha →hrasvaH i + e + itha →ecoya i + ay + itha →abhyAsasyA iyayitha "thou wentest"

KAZIKA iNo 'Ggasya yo 'bhyAsaH tasya dIrgho bhavati kiti liTi parataH. IyatuH, IyuH. iNo yaNN iti yaNAdeze kRte sthAnivadbhAvAd dvirvacanam. kiti iti kim? iyAya. iyayitha.

In the Iyatus example above, how did liTidhA reduplicate y into i + y?

Rule acaHparasmin makes that y be i-like enough to make the stammer that i would have made.

Why didn't iNoyaN work before Nal and ithal? They too start with vowels.

Because acoJNiti and hardsoft are stronger than iNoyaN.