1
SIMPLICITY, NOT SIMPLISM: IN DEFENCE OF OCCAM’S
RAZOR AS SCIENTIFIC NECESSITY
Au ho : A. Vodopyano
1
P e-p in a chi ed a SSRN Elec onic Jou nal: h p://doi.o g/10.2139/ss n.5570539
A chi ed a Zenodo (mi o ): h p://doi.o g/10.5281/zenodo.17261035
Indexed on PhilPape s: h ps://philpape s.o g/ ec/VODBFP
1ORCID: 0009-0007-1900-6073
Email:
wona u es@li e.com
Independen Resea che
2
Abs ac
Occam’s Razo (also known as he p inciple o pa simony and on ological educ ionism) is egula ly
dismissed as an o namen al es ige—no longe wo hy o i s longs anding s a us as a undamen al
scien i ic p inciple. An in e p e a ion he eo called Ve i ica ional Simplici y o Blind Fai h Pa simony is
o e ed, which: (a) iden i ies he mos scien i ic hypo hesis as ha yielding he ulles accoun o
co esponding da a while necessi a ing leas e i iabili y-a e se assump ion (b) unde lies all in eg al
scien i ic alues and (c) is o pa amoun scien i ic impo ance. Th ee suppo ing a gumen s a e employed.
A gumen 1 highligh s he o en-o e looked agili y wi h which Occam’s Razo acco ds o la es da a.
A gumen 2 shows ha wi hou pa simony o block ad hoc hypo heses, alsi iabili y is
impossible— ende ing science i sel un enable. While a gumen 3 co e s he eno mous bu den o p oo
easonable co esponding dismissals mus shoulde .
Keywo ds: Philosophy o Science, Pa simony, Ve i iabili y, Falsi iabili y, Reduc ionism, Bayesian
P obabilism
In oduc ion
As eloquen ly summa ized by McFadden (2023: 8, 11):
“Occam’s azo ... has ecen ly been a acked as a cul u al bias wi hou a ional ounda ion.
…
This c i icism o he in luence o Occam’s azo in science emains common. [I ] has been a acked wi h
claims ha ‘i s he o ical pu pose [is] as an old saw pe suading us o champion he supposed i ue o
simplici y.’
44
In sys ems biology, o example, i has been claimed ha because li e is ‘i educibly
complex’ Occam’s azo has no ole in model selec ion.
45
Recen popula science a icles by in luen ial
au ho s ha e made simila claims… a guing ha Occam’s azo ep esen s he ‘ y anny o simple
explana ions’
46
o ha i is ‘appealing, widely belie ed, and deeply misleading’.
47
… Likely
con ibu [ing] o wha has been desc ibed as ‘a wo ying end o a ou unnecessa ily complex
in e p e a ions’.
48
”
Ox o d Re e ence’s Dic iona y o Epidemiology desc ibes Occams Razo as:
“The p inciple o scien i ic pa simony (pa simony in he sense o unwillingness o use unnecessa y
esou ces, [which is also some imes e e ed o as scien i ic] ugali y, aus e i y). An ancien p inciple
o en a ibu ed o he philosophe … William o Ockham (c.1285-c.1349), who said:… assump ions o
explain a phenomenon mus no be mul iplied beyond necessi y.”
3
The Rou ledge ”P obabili y, Choice, and Reason” ex book by P o . Leigh on Vaughan Williams,
elabo a es hus (2021: 186, 189):
“Occam’s Razo has come o embody he me hod o elimina ing unnecessa y hypo heses. Essen ially, [i ]
holds ha he heo y which explains all (o he mos ) while assuming he leas is he mos likely o be
co ec . … This is he p inciple o pa simony. …
Occam’s Razo … poin s o he simples explana ion ha is consis en wi h he da a a ailable a a gi en
ime, bu e en so he simples explana ion may be uled ou as new da a become a ailable. This does no
in alida e he Razo , which does no s a e ha simple heo ies a e necessa ily mo e ue han mo e
complex heo ies, bu ha when mo e han one heo y explains he same da a, he simple should be
acco ded mo e p obabilis ic weigh .[
2
] So Occam’s Razo … is also consis en wi h mul iplying en i ies
which a e in ac necessa y o explain a phenomenon.” E.g., In sys ems biology.
Two ap opos co olla y p inciples a e known as:
a. Hi chens’s azo (2007: 150): “Wha [is] asse ed wi hou e idence can… be dismissed wi hou
e idence.”
b. Sagan s anda d (1979: 73): “Ex ao dina y claims equi e ex ao dina y e idence.”
3
As expanded-upon in A gumen 1 below, said da a/e idence consis ency poin is absolu ely c ucial.
A gumen 1: Da a Consis ency
The mos popula co esponding example is known as duck es . Dic iona y.com b ie ly elabo a es:
“I i looks like a duck, swims like a duck, and quacks like a duck, says Occam’s azo , i ’s p obably a
duck. No a goose disguised as a duck ha in il a ed he lock.”
This is so because he added con olu ion/ex ao dina iness o ha la e goose-in il a o hypo hesis,
p o es supe luous (and he e o e unscien i ic) when he ex ao dina y e idence needed o subs an ia e
i —is missing. Occam’s Razo compelling esea che s o p io i ize he mo e logically s aigh o wa d and
he e o e mo e scien i ic duck hypo hesis ins ead. As keenly obse ed by Williams abo e howe e , his
would cease o be he case i “new da a became a ailable” which e ealed he goose hypo hesis o be
mo e logically s aigh o wa d a e all. E.g., I he animal was ac ually e i ied o ake o i s duck
2
Such “p obabilis ic” applica ion o Occam’s Razo , quickly p o es a logical necessi y. E.g., Because ea ing
spec a as bina ies a he cos o c ucial co esponding da a is con a y o being “consis en wi h he da a a ailable.”
I.e., To yielding he ulles accoun he eo . Na u ally leading o ano he Occam’s Razo co olla y known as
Bayesian P obabilism/In e ence.
3
As T essoldi (2011: 1) no es, his p inciple is Sagan’s popula ewo ding o “Laplace’s p inciple[:] ‘ he weigh o
e idence o an ex ao dina y claim mus be p opo ioned o i s s angeness’ (Gillispie e al., 1999).”
4
cos ume; wi hou which i indeed looked, ac ed and sounded like a goose. In ligh o his ex ao dina y
new da a, i is hen he disguised-goose hypo hesis ha p o es he mo e logically s aigh o wa d, less
con olu ed and he e o e mo e scien i ic o he wo. Because i would hen be e en mo e con olu ed o
claim ha i was none heless a duck all along. Which would ha e had o success ully disguise i sel as a
goose who, in u n, disguised i sel as a duck. Unless e en mo e ex ao dina y e idence/da a came o ligh ,
which ende ed his e en mo e-ex ao dina y hypo hesis he mos s aigh o wa d, plausible op ion. And
so on.
In ei he case, Occam’s Razo indeed emains e e help ul and in ac . Simply eminding esea che s o
p io i ize whiche e hypo hesis ma ches he a ailable co esponding da a mos e i iably. Keeping he
ex en o blind assump ion equi ed o accep i , down o minimum and hus keeping said hypo hesis as
scien i ic as humanly possible. Which b ings us o he no ion ha in e p e ing Occam’s Razo me ely as
p e e ing supe icial simplici y—is simplis ic and e oneous. A mo e accu a e in e p e a ion o i s aim
being Ve i ica ional Simplici y. I.e., Blind Fai h Pa simony: he mos scien i ic hypo hesis is whiche e
yields ulles accoun o co esponding da a, while necessi a ing leas e i iabili y-a e se assump ion.
A c i ical dis inc ion ha is oo o en o e looked in a gumen s agains Occams Razo being scien i ically
undamen al.
A gumen 2: No Pa simony → No Falsi iabili y → No Science
The ollowing 3-s ep a gumen es ablishes co esponding undamen ali y u he ye .
S1. No Falsi iabili y = No Science
Despi e occasional challenges (o en in se ing ashionable heo ies which hemsel es inc easingly p o e
un alsi iable), he no ion ha alsi iabili y is cen al o iden i ying pseudoscien i ic heo ies—is well
ounded and es ablished in mains eam academia. As b ie ly demons a ed below.
Encyclopedia o Science and Religion’s alsi iabili y en y s a es:
“Falsi iabili y [is] he mos commonly in oked ‘c i e ion o dema ca ion’ o science om nonscience.”
McFadden (2023: 8) adds:
“Philosophy o science is a ely augh as a componen o scien i ic educa ion, bu i pushed o iden i y
he de ining ea u e o hei disciplines mos scien is s gene ally choose he p inciple o alsi iabili y
(a ibu able o Ka l Poppe ).
14
”
5
A p ominen example o co esponding con ex ualiza ion is Michael Ruse’s (Be and Russell Socie y
awa d) ellingly i led 1982 “C ea ion Science Is No Science”:
“Religion [and] c ea ion-science is no science [because genuine] science mus be open o change,
howe e con iden one may eel a p esen . Fana ical dogma ism is jus no accep able. … I he ac s
speak agains a heo y, hen i mus go. A [genuine] body o science mus be alsi iable.”
S an o d Encyclopedia o Philosophy expands upon he co esponding posi ion hus:
“I a heo y is incompa ible wi h possible empi ical obse a ions i is scien i ic; con e sely, a heo y
which is compa ible wi h all such obse a ions… is unscien i ic.”
Camb idge dic iona y adds:
“Falsi iable [means] able o be p o ed... alse. [E.g.,] All good science mus be alsi iable.”
Why alsi iabili y is so scien i ically-essen ial, is u he delinea ed ia Ox o d Re e ence. Acco ding o
which, science i sel is:
“The sys ema ic s udy o he s uc u e and beha io o he physical and na u al wo ld h ough
obse a ion, expe imen a ion, and he es ing o heo ies agains he e idence ob ained.” (As ci ed in
McHugh, 2024: 88)
E go, i a heo y canno be alsi ied when “ es [ed] agains e idence”, hen i simply isn’ genuinely
e idence-based and he e o e—isn’ scien i ic. Pa icula ly so, when es able al e na i es exis (as is
i ually always he case).
S2. No Pa simony = No Falsi iabili y
Pe Poppe (2008: 731):
“Falsi ica ion o a heo [y] can always be a oided by in oducing an auxilia y hypo hesis.”
These a e also e e ed o as ad hoc hypo heses, immunizing s a agems o ins ances o special pleading.
Schindle (2024: 71) elabo a es:
“When in oduced o sa e a heo y, ad hoc hypo heses would educe he heo y’s alsi iabili y, and
he e o e ‘deg ees o ad hoc-ness a e ela ed (in e sely) o deg ees[
4
] o es abili y and signi icance’
(Poppe , 1959). …
Ad hoc hypo heses a e hypo heses which may make independen p edic ions, bu o which he e [is] no
suppo . This iew is inc edibly popula … (Scha ne , 1974; Leplin, 1975; …). The ollowing example
4
Once again, i being a mo e accu a e o assess such ac o s in “deg ees” a he han in simplis ic bina y e ms.
6
is o en in oked o mo i a e he iew (e.g., Wo all, 2002): when an i egula i y was disco e ed in he
plane U anus, Adams and Le Ve ie in 1845-46 p oposed ha a new plane in he icini y o U anus
migh cause he disc epancy. A i s , he ‘Nep une hypo hesis’ was ad hoc, because i was in oduced o
sa e New on’s heo y. Bu a e Nep une was disco e ed, he hypo hesis was independen ly con i med
and hus los i s ad hoc s a us.
15
”
O as eloquen ly eph ased in he Wikipedia Occam’s Razo en y
5
:
“E en i some inc eases in complexi y a e some imes necessa y, he e s ill emains a jus i ied gene al
bias owa d he simple o wo compe ing explana ions. To unde s and why, conside ha o each
accep ed explana ion o a phenomenon, he e is always an in ini e numbe o possible, mo e complex,
and ul ima ely inco ec , al e na i es. This is so because one can always bu den a ailing explana ion
wi h an ad hoc hypo hesis.. p e en [ing i ] om being alsi ied. …
This endless supply o elabo a e compe ing explana ions… canno be echnically uled ou – excep by
using Occam’s azo .”
I.e., In lieu o he da a/e idence needed o de end a pa icula heo y, “one can always [p op i up] wi h an
ad hoc hypo hesis”. E.g., Dinosau ossils con adic he Biblical accoun , p omp ing some Ch is ians o
amously claim ha said ossils we e plan ed by God himsel o es humani y’s ai h. Which is ad hoc
because i elies on a deeply-selec i e su ey o co esponding da a, while necessi a ing a as
compa a i e measu e o e i iabili y-a e se assump ion. A maneu e ha is no mo e scien i ic han
in oking ha disguised-goose hypo hesis on li le o no e idence.
S3. E go: No Pa simony = No Science
Gi en he wo p eceding poin s, i should indeed be clea ha “endless… elabo a e compe ing [ad hoc]
hypo heses, canno be echnically uled ou – excep by using Occam’s azo ” in some o m. I.e., Tha
unless esea che s i s employ such pa simony o help weigh scien i ic p obabili ies agains one
ano he —all hypo heses become un alsi iable; and as such—unscien i ic. Lea ing no means o objec i ely
disce n mo e scien i ic, s aigh o wa d and likely explana ions om hei mo e unscien i ic, con olu ed
and unlikely kin; o dis inguish needless blind ai h om ha , as-ye scien i ically una oidable.
5
As wi h Dic iona y.com, Wikipedia is used he e as a e ia y sou ce due p ima ily o he soundness o i s ph asing.
The unde lying concep s ha ing been (a) es ablished elsewhe e in his pape ia bo h basic easoning and p ima y
sou ces, as well as (b) es ablished ia co esponding sou ces ci ed by Wikipedia i sel (e.g., S ano ich, 2007: 19–33;
Ca oll, 2008; Swinbu ne, 1997).
7
A gumen 3: Bu den O P oo
The Blind Fai h Pa simony in e p e a ion o Occam’s Razo mo eo e lea es c i ics o ackle ques ions
like:
a. A e cen u ies o s udy, can so much as a single example be clea ly iden i ied whe e abo e-de ailed
pi alls a e a oided and ye he co esponding p inciple ails? I.e., Is he e a single hypo hesis which yields
ulles accoun o co esponding da a while necessi a ing leas e i iabili y-a e se assump ion and ye
p o es less scien i ic han he al e na i e(s)? So p io o new co esponding da a becoming a ailable?
b. E en i such a case can be co obo a ed, can i easonably be cha ac e ized as no an excep ion ha
p o es he ule?
c. I he answe o ‘b’ (much less ‘a’) abo e is ‘no’—does his lea e any easonable oom o doub o e
whe he such pa simony is a co e scien i ic p inciple?
Co esponding dismissals need demons a e no only ha pu a i e excep ions a e genuine and signi ican ,
bu also:
i. Tha a iable al e na i e a oids compa a i ely mo e he eo .
ii. Tha his al e na i e ei he di e s in essence om said p inciple o is e en mo e undamen al.
O he wise i could no cons i u e a genuine al e na i e o Occam’s azo , bu only ano he co olla y
6
and/o complemen he e o a mos .
Conclusion
Only misin e p e a ion and misuse o Occam’s Razo p o es an i he ical o genuine science. Such as
con la ing he e i ica ional simplici y p e e ed he eby wi h simplism (e.g., igno ing ex ao dina y
coun e e idence so as o a i icially bols e simplis ic explana ions). Whe eas genuine Occam’s Razo
exp essly o bids such misuse, p o ing i sel essen ial o genuine science a e all.
6
I is mo eo e e en di icul o iden i y a single in eg al scien i ic alue which (in i s op imal o m) is no unde lain
by such pa simony. I.e., Which does no appea u he along he logical pa h ini ia ed he eby. F om alsi iabili y,
e i iabili y, Hi chens’s Razo , Sagan S anda d and Bayesian p obabili y as shown abo e, o o he such alues like
cohe ence, p edic i e powe and explana o y scope. A hypo hesis sco ing poo ly on ei he such on , in a iably
p o ing o be less capable o yielding a ull accoun o co esponding da a and/o o minimizing e i iabili y-a e se
assump ion.
8
Re e ences
Ca oll, R. T. (2008, June 22). Ad hoc hypo hesis. The Skep ic’s Dic iona y.
h ps://web.a chi e.o g/web/20251001215830/h ps://skepdic.com/adhoc.h ml
Dic iona y.com. (2018). Occam’s azo .
h ps://web.a chi e.o g/web/20251001203111/h ps://www.dic iona y.com/e/pop-cul u e/occams- azo /
Encyclopedia o Science and Religion. (2003). Falsi iabili y. Macmillan Re e ence USA.
h ps://web.a chi e.o g/web/20251001203814/h ps://www.encyclopedia.com/science-and- echnology/physics/science-gene al/ alsi iabili y
Hi chens, C. (2007). God is no g ea : How eligion poisons e e y hing. Twel e.
McFadden, J. (2023). Razo sha p: The ole o Occam’s azo in science. Annals o he New Yo k
Academy o Sciences, 1530, 8–17.
h ps://doi.o g/10.1111/nyas.15086. A ailable a
h ps://pmc.ncbi.nlm.nih.go /a icles/PMC10952609/
.
McHugh, K. (2024). A scien is ’s jou ney: lea ning o communica e science o imp o ed na u e
connec edness. The Li ing Lab, 1(1), 86–107.
h ps://doi.o g/10.20933/40000104
. A ailable a
h ps://web.a chi e.o g/web/20251001210254/h ps://disco e y.dundee.ac.uk/ws/po al iles/po al/140249122/5._McHugh_K._86-107_issue_1.pd
Ox o d Re e ence. (2017). Science and echnology. A ailable a
h ps://www.ox o d e e ence.com/page/134
(as
ci ed in McHugh, 2024: 88).
Poppe , K. (2008). The wo undamen al p oblems o he heo y o knowledge (T. E. Hansen, Ed.; A.
Pickel, T ans.). Rou ledge.
Po a, M. (2014). Occam’s Razo . In A Dic iona y o Epidemiology. Ox o d Uni e si y P ess.
h ps://web.a chi e.o g/web/20250926191322/h ps://www.ox o d e e ence.com/display/10.1093/ac e /9780199976720.001.0001/ac
e -9780199976720-e-1336? skey=ZOocoh& esul =1519
Ruse, M. (1982). C ea ion Science Is No Science. Science, Technology, &Human Values, 7(3), 72-78.
h ps://doi.o g/10.1177/016224398200700313
Sagan, C. (1979). B oca’s b ain: Re lec ions on he omance o science. Random House.
9
Schindle , S. (2024). P edic i ism and a oidance o ad hoc-ness: An empi ical s udy. S udies in His o y
and Philosophy o Science, 104, 68–77.
h ps://doi.o g/10.1016/j.shpsa.2023.11.008
S ano ich, K. E. (2007). How o Think S aigh Abou Psychology. Pea son Educa ion.
Swinbu ne, R. (1997). Simplici y as e idence o u h. Ma que e Uni e si y P ess.
Tho n on, S. (2023). Ka l Poppe . In E. N. Zal a & U. Nodelman (Eds.), The S an o d encyclopedia o
philosophy (Fall 2023 ed.). S an o d Uni e si y.
h ps://pla o.s an o d.edu/a chi es/ all2023/en ies/poppe
T essoldi, P. E. (2011). Ex ao dina y claims equi e ex ao dina y e idence: The case o non-local
pe cep ion, a classical and Bayesian e iew o e idences. F on ie s in Psychology, 2, A icle 117.
h ps://doi.o g/10.3389/ psyg.2011.00117
Wikipedia con ibu o s. (2025). Occam’s azo . In Wikipedia.
h ps://en.wikipedia.o g/w/index.php? i le=Occam%27s_ azo &oldid=1313302051
Williams, L. V. (2021). P obabili y, Choice, and Reason (1s ed.). Chapman and Hall/CRC.
h ps://doi.o g/10.1201/9781003083610
