0% found this document useful (0 votes)
383 views28 pagesCausation and Explanation in Social Science
Uploaded by
F. FCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
0% found this document useful (0 votes)
383 views28 pagesCausation and Explanation in Social Science
Uploaded by
F. FCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
CHAPTER 10
CAUSATION AND
EXPLANATION IN
SOCIAL SCIENCE
HENRY E. BRADY
1 CAUSALITY
Humans depend upon causation all the time to explain what has happened to them,
to make realistic predictions about what will happen, and to affect what happens in
the future, Not surprisingly, we are inveterate searchers after causes. Almost no one
goes through a day without uttering sentences of the form X caused ¥ of ¥ occurred
because of X. Causal statements explain events, allow predictions about the future, and
‘make it possible to take actions to affect the future. Knowing mote about causality can
bbe useful to social science researchers.
Philosophers and statisticians know something about causality, but entering into
the philosophical and statistical thickets is a daunting enterprise for social scientists
because it requires technical skills (e.g. knowledge of modal logic) and technical
information e.g, nowledge of probability theory) that isnot easily mastered. The net
payoff from forays into philosophy or statistics sometimes seems small compared to
the investment required. The goal of this chapter isto provide a user-friendly synopsis
of philosophical and statistical musings about causation. Some technical ises will
be discussed, but the goal will abyays be to ask about the bottom line—how ean this
information make us better researchers?
‘Three types of intellectual questions typically arse in philosophical discussions of
causality 28 WENRY E, BRADY
+ Psychological and lingustic—What do we mean by causality when we use the
‘concept?
‘+ Metaphysical or ontological—What is causality?
« Epistemological—How do we discover when causality is operative?”
Four distinct approaches to causality, summarized in Table 10.1, provide answers to
these and other questions about causality.” Philosophers debate which approach is,
the right one. For our purposes, we embrace them all. Our primary goal is developing
better socal science methods, and our perspective is that al these approaches capture
some aspect of causality, Therefore practical researchers can profit from drawing
lessons from each one of them even though their proponents sometimes treat them
as competing or even contradictory. Our standard has been whether or not we
‘could think of concrete examples of research that utilized (or could have utilized)
a perspective to some advantage. IT we could think of such examples, then we think it
is worth drawing lessons from that approach.
‘A really good causal inference should satisly the requirements of all four ap-
proaches. Causal inferences willbe stronger to the extent that they are based upon
finding all the following: (1) Constant conjunction of causes and effects required by
the neo-Humean approach. (2} No effect when the cause is absent in the most similar
‘world to where the cause i present as required by the counterfactual approach. (3) An
effect after a cause is manipulated. (4) Activities and processes linking causes and
‘effects required by the mechanism approach.
‘The claim that smoking causes lung cancer, for example first arose in epidemiolog:
ical studies that found a correlation between smoking and lung cancer. These results
were highly suggestive to many, but this correlational evidence was insufficient to
‘others (ineluding one ofthe founders of modern statistics. A Fisher). These studies
‘were followed by experiments that showed that, a least in animals, the absence of
smoking reduced the incidence of cancer compared to the incidence with smoking
‘when similar groups were compared. But animals, some suggested, are not people.
(Other studies shoved that when people stopped smoking (that is, when the putative
‘cause of cancer was manipulated) the incidence of cancer went down as well. Finally,
recent studies have uncovered biological mechanisms that explain the link between
smoking and lung cancer. Taken together the evidence for a relationship between
smoking and lung cancer now seems overwhelming.
"fourth question pragmatic How do we convince other to acep our explanation o asa
argument? leading proponent of this approach Ba van Fans ho). Kater and Salon
{ip is] ange that "van rasan se fered the bet theory of the pragmatic if explanation to ate,
‘atthe seen ak eagmatic theory of explanation then faces serious dicate”
[ecoe there ference betwen "a theory ofthe pragma of explanation ada pragmatic theory
‘feaplamation” From es perpetvg, knowing hot peopl convince thers of heory does Bot se
the antago epstemologal procs.
Fw important ols om causality are ot covered inthis chapter although the author has profited
from thet insights Per (200) provide comprehensive approach to cay rote ina Bayesian
perspective Shafer (top) link decison theory a causal resin a nove and etl wy.
‘Table 10. Four approaches to causality
Q
a
g
g
f
Countertacuat
Neo-Humean equa
Harr and Madden (1975)
Casting (1955); Mere and
z
i
}
&
Price (1983);
von Wright (1971)
§
i
Beauchamp and
Rosenberg (198)
Dbseration of constant
Consideration of whether
Recipe that regulary
in otherwise sar
Tha
theresa mecharism or
produces the effect from
the cause
‘urjunetion and
i
i
5
:
i
i
a
4
i
g
i
capacity that leads fom
the cave to the eect,
nif the cause
effect
g
8
a
BE
be
is
u
i]
‘An appeal to the operation of
the metharism
Observation ofthe eet of
‘the maipuaion|
i
Temporal precedence
-Anproach tote asymmetric
spect of esa
:
3
2
é
g
a
i
Z
direction
Effects of eauses
q
3
é
=
§
necesity
Sing
Eefects of causes
Necessary connection
Major problens solved
Causes of eects
{eg focus on mechaism
that create effects)
periments)
(eg focus on teatment’s
sitet
experiments:
(eg focs on treatment’
effets in experiments)
epetimen
(eg focus on dependent
‘variable in eoresios)
observational and cause
causes of eects
fects or eects of
causes?
Stusies with comparative
Emphasis on causes of
lemodels cae studies
natural experiments;
(uas-experiments
‘ise study comparisons;
‘counterfactual thougnt
experiments
‘modeling
advantage sing this
Sefton
219 2 COUNTERFACTUALS
Causal statements are so useful that most people cannot let an event go by without
asking why it happened and offering their own “because” They often enliven these
“discussions with counterfactual asertions such as “if the cause had not occurred,
then the effet would not have happened.” A counterfactual isa statement, typically
in the subjunctive mood, in which a false or “counter to fact” premise is followed
by some assertion about what would have happened if the premise were true. For
example, the butterfly ballot was used in Palm Beach County Florida in 2000 and
{George W. Bush was elected president. A counterfactual assertion might be “ifthe
patterfly ballot had not been used in Palm Beach County in 2000, then George
Buch would not have been elected president.” The statement uses the subjunctive
(Cif the butterfly ballot had not been used, ..then George Bush would not have
‘been elected”), and the premise is counter to the facts. The premise is false be-
cause the butterfly ballot was used in Palm Beach County in the real world as it
‘unfolded, The counterfactual claim is that without this ballot, the world would
have proceeded differently, and George Bush would not have been president. Is this
true?
“The truth of counterfactuals is closely related to the existence of causal relation
ships, The counterfactual claim made above implies that there isa causal ink between
the butterfly ballot (the cause X) and the election of George Bush (the effect ¥). The
counterfactual, for example, would be true if the butterfly ballot aaused Al Gore to
ose enough votes so that Bush was elected. Then, ifthe butterfly ballot had not been
used, Al Gore would have gotten more votes and won the election.
“Another way to think about this isto simply ask what would have happened in
the most similar world in which the butterfly ballot was not used. Would George
Bush still be president? One way to do this would be to rerun the world with the
‘cause eradicated so thatthe butterfly ballot was not used. The world would otherwise
be the same. If George Bush did not become president, then we would say that
the counterfactual is true. Thus, the statement that the butterfly ballot caused the
‘lection of George W. Bush is essentially the same as saying that in the mos similar
svorld in which the butterfly ballot did not exist, George Bush would have lost, The
Uxistence ofa causal connection can be checked by determining whether oF not the
Counterfactual would be true in the most similar possible world where its premise is
tue, The problem, ofcourse is defining the most similar world and finding evidence
for what would happen init
‘Beyond these definitional questions about most similar worlds, there isthe prob
Jem of finding evidence for what would happen in the most similar world. We eannot
ferun the world 30 that the butterfly ballot is not used. What can we do? Many
‘Philosophers have wrestled with this question, and we discuss the problem in detail
Tater in the section on the counterfactual approach to causation.’ For now, we merely
> standard theories fli cana and counteracts becuse propositions with file remiss
ae ered trv which wold ean that alleourerfital atemens withthe fle
note that people act as if they can solve this problem because they assert the truth of
counterfactual statements all the time,
3 EXpLorinG THREE BASIC QUESTIONS
ABOUT CAUSALITY
‘Causality is atthe center of explanation and understanding, but what, exactly, isi
[And how isit related to counterfactual thinking? Somewhat confusingly, philosophers
‘mingle psychological, ontological, and epistemological arguments when they discuss
causality. Those not alerted to the different purposes of these arguments may find
philosophical discussions perplexing as they move from one kind of discussion to
lanothee, Our primary focus is epistemological. We want to know when causality is,
truly operative, not just when some psychological process leads people to believe that
itis operative. And we do not care much about metaphysical questions regarding,
‘what causality really is, although such ontological considerations become interesting
to the extent that they might help us discover causal relationships.
3a Psychological and Linguistic Analysis
Although our primary focus is epistemological, our everyday understanding, and
‘even our philosophical understanding, of causality is rooted in the psychology of
causal inference, Perhaps the most famous psychological analysis is David Hume’
investigation of what people mean when they refer to causes and effects. Hume (171~
76) was writing ata time when the pre-eminent theory of causality was the existence
ff a necessary connection—a kind of “hook” or “force’—between causes and their
effects so that a particular cause must be followed by a specific eect. Hume looked
for the feature of causes that guaranteed their effects. He argued that there was nO
evidence for the necessity of causes because all we could ever find in events was the
‘contiguity, precedence, and regularity of cause and effect. There was no evidence for
any kind of hook or force. He described his investigations as follows in his Treatise of
Huan Nature (1739)
Whats ouridea of neces, when wesay that wo objets are necessarily connected together...
Consider in what objects necessity is commonly supposed to le; and finding that iis always
Sseribed to causes and effects I turn my eye to two objects supposed to be placed in that
remiss, would be tue regrdesof whether of not caus ink existe Moda ais which ty
‘lpr the nature neces pos, contingency. and nspossbiity, have ben developed or
Caen tN ee aby eigaatretnes ofthe commer
‘ete or not the statement wold etre in the most sar poste world where the premise is
true Prats aris, however ia defining the mos sar work relation, and examine them in all he stations of which they are susceptible 1 immediatly
perceive that they are contiguous in time and place, and that the object we call cause precedes
the other we cal effect. Inno one instance can Igo any further, nor is it possible for me to
‘discover any third relation betwixt these objets, [therefore enlarge my view to comprehend
Sceral instances, where | find ike objects alway existing in like relations of comtigoity and
“secession. The reflection on several instances only repeats the same object; and therefore can
never give rise tow new idea. But upon father inguiy ind that the repetition is notin every
particular the same, but produces a new impression, and by that means te idea which Vat
present examine, Fo, after a fequeat repetition | fnd that upon the appearance of one ofthe
[bjects the mind is determine by custom to consider its usual atendant, an to consider it in
‘stronger light upon acount oft relation to the fist object. Its this impression, then, or
determination, which affords me the dea of necesty. (Hume, 1978 1730155)"
“Thus for Hume the idea of necessary connection is a psychological trick played by
the mind that observes repetitions of causes followed by effects and then presumes
‘some connection that goes beyond that regularity. For Hume, the major feature of
‘causation, beyond temporal precedence and contiguity, is simply the regularity ofthe
association of causes with ther effects, but there is no evidence for any kind of hook
for necessary connection between causes and effects
TThe Humean analysis of causation became the predominant perspective in the
‘nineteenth and most of the twentieth century, and it led in two directions, both of
Which focused upon the logical form of causal statements, Some, such asthe physicist
Ernst Mach, the philosopher Bertrand Russell, and the statistician/geneticist Karl
Pearson, concluded that there was nothing more to causation than regularity o that
the entire concept should be abandoned in favor of functional laws oF measures of
association such as correlation which suramarized the reglarity.® Others, such asthe
philosophers John Stuart Mil (1888), Karl Hempel (1965), and Tom Beauchamp and
* tne gui a 44) which eter reworking ofthe Tete, Hume say" tha pon the
nol there appears ot toot all nts, any on instance of coaneion, which sconciabe by
{Gc Alleen seem entieyoose and separate One event alos ameter but we never nerve ary
tye betwee them They seem cone, bu ever comeced. And as we can haven ide of ting,
Uthich never apeated to ou outward sen or mar sentient the necessary concson seems tobe,
Utne hve mesa connenom os power atl and hat these words are bsoltely without meaning,
(nen cmployed either in psp esting or common fe. Thiscanexion, therefore we fs
{iene mind thiscastomory transition of the naginaton fom one objet ots usta attendan, ithe
Sentiment or imprsson, fm which we frm the ies of ower or necessary connexion”
ether ar difere ncrpecationsf what Hunt mean Fors thorough dictions Bexuchamp
au Rosenberg (i).
+ Perera Ral famously wrote hat “the word casei inextricably ound up with mideading
stator as to mae comple extoson fom te pxapil vocabulary desi ..The iw of
‘tality lke so ch tht pases ster among posopbes sarc ofa bygone a, surviving ie
‘he monarchy. oly Decwse ts evroncay supposed to do no harm” (Rose 8). Karl easos
MeN cuton and rpc encore sReynd cach cared ndarentals a ater
fone hess another ei ait th imcratable arama of even sem since, nel the
egy case an eft Isthicatory anything bt conceptual init to experienc, and without
Sry ecs in pecepton yon statisti approximation? (Person ti). Tt this conception of
‘eationfetecen wo occurences emracing al eltonsip from absalute independence
‘cpl deyenience, which the wider category by wick we ave o replace thei ies of ansaion”
(Penson 057)
Alexander Rosenberg (1981), looked for ways to strengthen the regularity condition
s0 as to go beyond mere accidental regularities. For them, true cause and effect
regularities must be unconditional and follow from some lawlike statement. Their
‘neo-Humean approach improved upon Hume’s approach, but as we shal se, there
appears to be no way to define lawlike statements in a way that captures all that we
‘mean by causality.
‘What, then, do we typically mean by causality? In their analysis ofthe fandamental
‘metaphors used to mark the operation of causality, the linguist George Lakoff and
the philosopher Mark Johnson (19804; 1980b; 1999) describe prototypical causation
as “the manipulation of objects by force, the volitional use of bodily force to change
something physically by direct contact in one’s immediate environment” (1999, 177)
Causes bring, throws, hurl, propel, lead, drag, pull, push, drive, tear, theust, or fling
the world into new circumstances. These verbs suggest that causation is forced move-
‘ment, and for Lakoff and Johnson the “Causation 1s Forced Movement metaphor
is in a crucial way constitutive of the concept of causation” (187). Causation as
forceful manipulation difers significantly from causation as the regularity of cause
and effect because forceful manipulation emphasizes intervention, agency, and the
possibility thatthe fue to engage in the manipulation will prevent the effect from
happening, For Lakoff and Jonson, causes are forces and capacities that entail their
effects in ways that go beyond mere regularity and that are reminiscent ofthe causal
“hooks” rejected by Hume, although instead of hooks they emphasize manipula
‘mechanisms, forces, and capacities.”
“Causation as regularity” and “causation as manipulation” are quite diferent no-
tions, but each carries with it some essential features of causality. And each isthe basis
fora different philosophical or everyday understanding of causality. From a psycho-
logical perspective, ther differences emerge clearly in research done in the last fifteen
years on the relationship between causal and counterfactual thinking (Spellman a
‘Mandel 1999). Research on this topic demonstrates that people focus on diferent fac-
tots when they think causally than when they think counterfactualy. In experiments,
people have been asked to consider causal attributions and counterfactual possibilities
in car accidents in which they imagine that they chose a new route to drive home
and were hit by a drunk driver. People’s causal attributions for these accidents tend
to “focus on antecedents that general knowledge suggest would covary with, and
therefore predict, the outcome (eg, the drunk driver,” but counterfactual thinking
focuses on controllable antecedents such a the choice of route (Spellman and Mandel
1999, 123). Roughly speaking, causal attributions are based upon a regularity approach
to causation while counterfactual thinking is based upon a manipulation approach,
to causation. The regularity approach suggests that drunken drivers typically cause
accidents but the counterfactual approach suggests that in this instance the person's
Axe shal how wo diferent approaches to causation are confated here One approach
‘mphaszs agency and manipulation. The oer approach emphasizes mechanisms and capaci. The
major ference the locus ofthe unerying force tat defines caus lationship. Agency and
iinutn pecs nghihunan reno, Mechs an cape aro
‘emphas process within nature sl a4 HENRY E, BRADY fet
choice of a new route was the cause of the accident because it was manipulable by
the person, The logic of causal and the logic of counterfactual thinking are so closely
related that these psychological differences in atributions lead to the suspicion that
both the regularity and the manipulation approaches tell us something important,
bout causation,
3.2 Ontological Questions
Knowing how most people think and talk about causality i useful, but we are ul
timately more interested in knowing what causality actually is and how we would
discover it in the world. These are respectively ontological and epistemological ques-
tions.* Ontological questions ask about the characteristics ofthe abstract entities that
exist in the world, The study of causality raises a numnber of fundamental ontological
‘questions regarding the things that are causally related and the nature of the causal
relation?
What are the things, the “causes” and the “effects” that are linked by causation?
‘Whatever they are, they must be the same things because causes can also be effects
and vice versa. But what are they? Are they facts, properties, events, or something,
else? The practicing researcher cannot ignore questions about the definition of
events. One of the things that researchers must consider is the proper definition of
fan event," and a great deal of the effort in doing empirical work is defining events
suitably Not surprisingly tremendous effort has gone into defining wars, revolutions,
fiems, organizations, democracies, religions, participatory ats, political campaigns,
and many other kinds of events and structures that matter for social science re-
search, Much could be said about defining events, but we shall only emphasize that
defining events in a useful fashion is one of the major tasks of good social science
research,
‘A second basic set of ontological questions concern the nature ofthe causal rela-
tionship. Is causality different when it deals with physical phenomena (e.g. billiard
* rong speaking, pitsophy is concerned with thc kind of qvesions regarding "what s*
{ont how ita he ko” epsterolog, ana “what vale tas (thes and esti). In
"wen these questions tment century ilosphy ie lo pid a ret deaf stnton to
logical insti and een pcg als
Symbolic we om tik othe cas eatin ab satement XY whee [Link] Yi
sfc and sacl elation. X and ¥arethe things that ate causly elated apd isthe casa
relation Ax hls tr thi elatonsip usually considered to be incomplete (not all X and Y
‘recall related), ssymmetri forthe evens stare easly lated (ether XY YX bt not
‘oth, and tea [eX isnot possible)
1 vent alata in space and tine (“the WWE peace setlement at Verses”) but ts are
rot Theft that the WW I peace setlement waa Verses) For dicasons of causal and events
See Henne (i) ad forcast ad facts se Melos (9. Many posers refer to speacof|
ropes" which re particule properties Bhring 97 Somepisoper eet he we tht the
vworkl can be described in terms fein events ores and angie for vet ar ending things
(tere and Maden os ch 6
13, pothour of cations hat del wth the dfinon of vents and sca proces ae Abbott
(9:93 1995), Plea (204, Riker (97, Ty (8).
CAUSATION AND EXPLANATION 235,
balls hitting one another or planets going around stars) than when it deals with social
phenomena (democratization, business cycles, cultural change, elections) that are
Socially constructed?" What role do human agency and mental events play in causa-
tion?” What can we say about the time structure and nature of causal processes?" Our
attitude is that socialsciences about the Formation of concepts and the identification
of causal mechanisms. We believe that social phenomena such as the Protestant
etic, the system of nation states, and culture exist and have causal implications. We
also believe that reasons, perceptions, beliefs, and attitudes affect human behavior.
Furthermore, we believe that these things can be observed and measured,
“Another basic question about the causal relation is whether it is deterministic or
probabilistic. The classic model of causation is the deterministic, clockwork New-
tonian universe in which the same initial conditions inevitably produce the same
‘outcome. But modem science has produced many examples where causal relation.
ships appear to be probabilistic. The most famous is quantum mechanics where the
position and momentum of particles is represented by probability distributions, but
many other sciences rely upon probabilistic relationships. Geneticist, for example, do
‘not expect that couples in which all the men have the same height and all the women
have the same height will have children of the same height. In this case, the same set
of (observed) causal factors produce a probability distribution over possible heights.
‘We now know that even detailed knowledge ofthe couple's DNA would not lead to
‘exact predictions. Probabilistic causation, therefore, seems possible in the physical
sciences, common in the biological sciences, and pervasive in the socal sciences.
Nevertheless, following the custom of a great deal of philosophical work, we shall
start with a discussion of deterministic causation in order not to complicate the
analysis.
3.3 Epistemological Questions
Epistemology is concerned with how we can obtain intellectually certain knowledge
(what the Greeks called “episteme”). How do we figure out that X really caused Y? At
the dinner table, our admonition not to reach across the table might be met with
“I dido’t break the glass, the table shook,” suggesting that our causal explanation
for the broken glass was wrong. How do we proceed in this situation? We would
probably try o rule out alternatives by investigating whether someone shook the
table, whether there was an earthquake, or something else happened to disturb the
lass. The problem here is that there are many possibilities that must be ruled out,
and what must be ruled out depends, to some extent, on our definition of causality
oe epesemtte dlscsions se Duke 1, Berger and chan (98), von Weight
(7, See 7), Wen (99)
See ithey (9), va Wright (7, ch, Davidson (ab), Seal (69, Wend (999).
In vied ae of metaphors Peon ano) compares diferent Kinds of wc cence proces
‘wth tomas, earhguas, large meteorite, ad lal warning terms fhe tine haz fhe
‘ure andthe time hori ofthe impact He show that the oa proce i each ution ae qe
‘iron 226 _WENRY [Link]
Learning about causality, then, requires that we know what itis and that we know
hhow to recognize it when we see it. The simple Humean approach appears to solve
both problems at once, Two events are causally related when they are contiguous, one
‘recedes another, and they occur regularly in constant conjunction with one another.
‘Once we have checked these conditions, we know that we have a causal connection
But upon examination, these conditions are not enough for causality because we
would not say that night causes day, even though day and night are contiguous,
night precedes day, and day and night are regulary associated. Furthermore, simple
regularities like this do not make it easy to distinguish cause from effect—aftr all,
day precedes night as well as night preceding day so that we could just as wel, and
just as mistakenly, say that day causes night. Something more is needed. 1¢ is this
something more that causes most ofthe problems for understanding causation. John.
Stuart Mill suggested that there had to be an “unconditional” relationship between
‘cause and effect and modern neo-Humeans have required a “lawlike” relationship,
but even if we knovr what this means" (which would solve the ontological problem
‘of causation) it is hard to ensure that it is true in particular instances so as to solve
the epistemological problem.
In the following sections, we begin with a review of four approaches of what
causality might be, We spend most of our time on a counterfactual definition, mostly
mounting toa recipe that is now widely used in statistics. We end with a discussion
‘of the imitations ofthe ecipe and how frit goes toward solving the epistemological
and ontological problems.
4 THe HUMEAN AND NEO-HUMEAN
APPROACH TO CAUSATION
4a. Lawlike Generalities and the Humean Regularity
‘Approach to Causation
Humean and neo-Humean approaches propose logical conditions that must hold for
the constant conjunction ofevens to justify the inference that they havea cause-effect
relationship. Specifically, Humeans have explored whether a cause must be sulficent
foritseflects necessary fr its effects or something more complicated.
» Something diferent might alo he needed Hume himself dropped he requirement fr contiguity
inhi ab rewnte of his work, and many posers woulda drops equrement foe
temporal precedes,
Ty tte newt this erature are presented with many statement about the pee for lawflness and
unconionalty whch se promie + ese tat wl ise nes. Bat the conitions hat ae
rete vay set lho of the goal
CAUSATION AND EXPLANATION 227,
caw —>
E:Burning Bung
oar —>»
Fig. 104, Two sts of INUS conditions
The classic definition shared by Hume, John Stuart Mill, and many others was
that “X is a cause of ¥ if and only if X is sufficient for ¥ That is the cause must
always and invariably lead to the effet. Certainly an X tha is sufficient for Y can be
considered a cause, but what about the many putative causes that are not sufficient
for their effect? Striking a match, for example, may be necessary for itt light, but it
‘may not light unless thete is enough oxygen in the atmosphere. Is striking a match
never a cause ofa match lighting? This leads to an alternative definition in which "X
isa cause of Y ifand only if X is necessary for ¥” Under ths definition, itis assumed
that the cause (such as striking the match) must be present for the effect to occur,
bbutit may not always be enough forthe cause to actually occur (because there might
not be enough oxygen). But how many causes are even necessary for their effects?
Ifthe match does not light after striking it, someone might use a blovetorch to light
it so that striking the match is not even necessary for the match to ignite. Do we
therefore assume that striking the match is never a cause ofits lighting? Necessity and
sufficiency seem unequal tothe tak of defining causation.”
‘These considerations led John Mackie to propose a set of conditions requiring that
a cause be an insufficient [I] bt necessary [N] part of condition which is itself un-
necessary (U] but exclusively suficent [S] forthe effect. These INUS conditions can
bbe explained by an example. Consider two ways that the effect (E), which isa building,
burning down, might occur (see Figure 10.1) In one scenario the wiring might short-
circuit and overheat, thus causing the wooden framing to bur. In another, gas
can might be next toa furnace that ignites and causes the gasoline can to explode.
‘A number of factors here are INUS conditions for the building to burn down. The
short circuit (C) and the wooden framing (W) together might cause the building to
‘burn down, or the gasoline can (G) and the furnace (F) might cause the building to
‘burn down. Thus, C and W together are exclusively suficient [S] to burn the building
down, and G and P together are exclusively sufficient [S] to burn the building down,
Furthermore, the short circuit and wooden framing (C and W) are unnecessary [U],
and the gasoline can and the furnace (G and F) are unnecessary [U] because the
‘building could have burned down with just one or the other combination of factors.
Finally, G W, G, or F alone is insufficient [1] to burn the building down even though
C is necessary [N] in conjunction with W (or vice versa) and G is necessary [N] in
conjunction with F (or vice-versa). This formulation allows forthe fact that no single
‘cause Is sufficient or necessary, but when experts say that a short clrcult caused the
"And hee ate problems chs the flowing favorite ofthe phsophers Iwo ballets piece
mans eat simultane fe esonable ose that each an ent part of tinct
facet condition ofthe det, and hat ether let str paris necsary forthe death, ince in
‘ch ce theater ult erent (Sot and Tok 9533) 228 WENRY [Link]
fire they “are saying in effec thatthe short-circuit (C) is condition of this sot that
iv occured, thatthe other conditions (W) which, conjoined with it form a sfiient
condition were also present, and that no other sufficient condition (uch as G and F)
Of the house's catching fre was present on this occasion” (Mackie 1965, 245 letters
addded).
From the perspective of practicing researcher, thee lessons follow from the INUS
conditions, First a putative cause such as C might not cause the effect F because G
‘and F might be responsible. Hence, the burned-doven building (E) will not always
result from a shor circuit (C) even though C could cause the building to bur dove.
Second, interactions among causes maybe necessary fr any one cause tobe sufficient
(C and W requir each ther and W and G require each other), Third, the relation
ship between any INUS cause and ts efect might appear to be probabilistic because
‘ofthe other INUS causes, In summary, the INUS conditions suggest the multiplicity
‘of causal pathways and causes, the possibility of conjunctural causation (Ragin 987),
and the likelihood that social science relationships will appear probabilistic even if
they are deterministic."
"A specific example might help to make these points clearer, Assume that the four
INUS factors mentioned above, C, W, Gand F, occur independently of one another
and that they are the only factors which cause firs in buildings. Further assume
that shor circuits (C) occur 10 percent ofthe time, wooden (W) frame buildings
50 percent ofthe time, farnaces (F) 90 percent ofthe time, and gasoline (G) cans
near furnaces 1 percent of the time. Because these evens are assumed independent
‘fone another, itis easy to calculate that Cand W occurs percent of the time and that
G and F occur 9 percent ofthe time. (We simply multiply the probability ofthe two
independent events) All four conditions occur 0.45 percent ofthe time. (The product
‘of al four percentages.) Thus, fires occur 135 percent of the time. Ths percentage
includes the cases where the ie isthe result of Cand W (5 percent ofthe ime) and
where it isthe result of G and F (9 percent of the time), and it adjusts downward
for double-counting that occurs inthe cases whet all four INUS conditions occur
together (0-45 percent ofthe time)
‘Now suppose an experimenter did not know about the role of wooden frame
‘buildings or gasoline cans and furnaces and only looked atthe relationship between
fires and short circuits. Across-tabulation of ies with the short circuit factor would
yield Table 10.2. As assumed above, short circuits occur ro percent of the time (see
the third column total atthe bottom ofthe table) and as calculated above, fires occur
1355 percent ofthe time (ee the third row total on the far right) The entries in the
interior ofthe table are ealulated ina similar way.”
ven though each case occurs because of a deterministic process—either a short
circuit and a wooden feame building or gasoline can and a furnace (or both) —
this cros-tabulation suggests a probabilistic relationship between fines and short
"These points are made espe ocflly in Marin nd Sige (95)
"Thus en for short ccs and es ores rom he ss where ere are short iris md
wooden fame buildings (3 psent ofthe time) andere thee re short cits apd no woolen Fame
Thing ut there are gasoline cans and furmaces (5 ptcent ines 9 pee).
NATION 229)
Table vo.a. Fees by short circu
percentages of each event)
in hypothetical example (otal
Not C=ne short circuits C—short eels Row totals
Not Eno fires 1.90 455 2648
Efires 8.10 545 1355
Coluen tras 90.00 10.00 100.00
Circuits. In 455 percent of the cases, short circuits occur but no fires result because
the building was not wooden. In 8.10 percent ofthe cases, there are no short circuits,
but a fire occurs because the gasoline can has been placed near the furnace. For
this table, a standard measure of association, the Pearson correlation, between the
effect and the cause is about 40 which is far short ofthe 1.0 required for a perfect
(positive) relationship. If, however, the correct model is considered in which there
are the required interaction effects, the relationship will produce a perfect fit.°® Thus,
‘a misspecification of a deterministic relationship can easily lead a researcher to think.
that there isa probabilistic relationship between the cause and effect.
INUS conditions reveal alot about the complexities of causality but asa definition
of it, they turn out to be too weak—they do not rule out situations where there
are common causes, and they do not exclude accidental regularities. The problem
‘of common cause arises in a situation where, for example, lightning strikes (L) the
‘wooden framing (WW) and causes it to burn (E) while also causing a short in the
circuitry (C). That is, L > Band + C (where the arrow indicates causation). IF
lightning always causes a short in the circuitry, but the short never has anything to
cdo with a fire in these situations because the lightning starts the ire directly throtgh
its heating ofthe wood, we will nevertheless always find that C and F are constantly
conjoined through the action of the lightning, suggesting thatthe short circuit caused.
the fie even though the truth is that lighting is the common cause of both.» In some
‘eases of common causes such as the rise in barometric pressure followed by the arrival
‘of storm, common sense tells us that the putative cause (the rise in barometric
pressure) cannot be the real cause of the thunderstorm. But in the situation with
the lightning, the fact that short circuits have the capacity to cause fires makes it
less likely that we will realize that lightning is the common cause of both the short
circuits and the fires. We might be better off in the case where the lightning split,
some ofthe wood framing of the house instead of causing short circuit In that ease,
teach variables scored 20 or one depending upon whether he eect ot cause i present or
atsent then a eession equation af the effect onthe product (or interaction of and W, the podct
fo'¢ and Fy andthe product ofc, W, Gand F wal produce a maltipe conteation of one indeang 2
perf ft
2 esa posta he lightning’ eating of the wood i abays o sometines) inset 10
‘cause their not L—» E),but ts cretion of short cei (L-» Calas or sone)
‘lie for thee (C —> Et ease, the hing i thease othe oe aug ts
eation oftheshor creat Thats, CB 230 HENRY E. BRADY
‘we would probably reject the fantastic theory that split wood caused the fire because
split wood does not have the capacity to start afte, but the Humean approach would
bbe equally confused by both situations because it could not appeal, within the ambit
ofits understanding, to causal capacities, For a Humean, the constant conjunction of
split wood and fires suggests causation as much as the constant conjunction of short
Circuits and fires. Indeed, the constant conjunction of storks and babies would be
treated as probative of a causal connection.
“Attempts to fix up these conditions usually focus on trying to require “awlike”
statements that are unconditionally true, not just accidentally true. Since itis not
"unconditionally true that splitting wood causes fires, the presumption is that some
‘such conditions can be found to rule out this explanation. Unfortunately, no set
‘of conditions seem to be successful.” Although the regularity approach identifies a
‘necessary condition for deseribing causation, it basically fails because association is,
rot causation and there is no reason why purely logical restrictions on lawlike state
‘ments should be sulicient to characterize causal relationships Part ofthe problem is
that there are many different types of causal laws and they do not fit any particular
patterns. For example, one restriction that has been proposed to ensure lawfulness is
that lawlike statements should either not refer to particular situations or they should
be derivable from laws that do not refer to particular situations. This would mean
that Kepler's fist “lve” about all planets moving in eliptcal orbits around the sun (a
highly specific situation!) was not a causal law before Newton's laws were discovered,
bout it was a causal lv after it was shown that it could be derived from Newton's laws.
‘But Kepler’ laws were aways considered causal laws, and there seems tobe no reason.
to rest their lawfulness on Newton's laws, Furthermore, by this standard, almost
all social science and natural science laves (eg. plate tectonics) are about particular
situations. In short, logical restrictions on the form of laws do not seem sufficient to
characterize causality,
4.2. The Asymmetry of Causation
“The regularity approach also fails because it does not provide an explanation for the
asymmetry of causation. Causes should cause their effects, but INUS conditions are
llmost always symmetrical such that if C is an INUS cause of E, then E is also an
INUS cause of C. Itis almost always possible to turn around an INUS condition so
that an effect is an INUS for its cause.* One of the most famous examples of this
problem involves a flagpole, the elevation ofthe sun, and the flagpole’s shadow. The
2 For some representa discussions ofthe problems se Harre and Madden 75h. 2): Salon
(oyporshn ss Flweman (och 3) Some gy) ate that Tae dal import whats
pocesay pose or po, and suppor of counteract seems to hae common extension
Satements er poses al cor aka tree Butts externa dificult find crits to
‘epurt the statements that do fo thos tht donot
1 Pape 29) provide demonstration ofthe symmetry of INUS conditions, and be wes
‘onto sat condition for he ayomatry of sation hat does nt lyon the empl latonstip
Tetween causes and fle.
lta ight ar in stright ins impli tat et tinh been the
tei of the ago, the eng ft sad andthe ange of eatin a te an
‘When te san nes the haw is lon at iy sot and tn is
tv Ton dt cvs ngs the th oft soe comedy
IB ih ofthe gpl andthe craton ofthe un Ba tsingINUSconlons,
wea swell sp tn the clean of he sun cused y the ght ofthe
fgets ong of coe Them htt cn ht
reine this atti pouty.
“Theor! atre ofc Haman approach that provides for aymetysteporl
ycadenc ange nth deafon ofthe aur pee arepentig cheng
Inte length of the hadom then we cn sy that the eonton of the sn case
the Ing of the shadow, An change a the bit ofthe nga freed
Corrependng change inthe length ofthe shadow, we can sy th the Bh of
the agp cae the nth of the shadow, But may pilseprs et making
tempor precedence the eemiaant of oul eyeaty Bere preheat
posi exlaning the dectn of ine by cas ayramety an pecs
the pssbility of baclrards auton, Prom a praca perspective al oie
cas ca ong th abe cl nx prt ct
43 Summary
‘This discussion reveals two basic aspects of the causal relation. One is asymmetrical
form of association between cause and effet and the other isan asymmetrical relation
in which causes produce effects but not the reverse. The Humean regularity approach,
in the form of INUS conditions, provides a necessary condition forthe existence of
the symmetrical relationship, but it does not rule out situations such as common
‘cause and accidental regularities where there is no causal relationship at all. From
‘a methodological standpoint, it can easily lead researchers to presume that all they
reed to do is to find associations, and it also leads to an underemphasis on the
rest of the requirement for a “Iawlike” or “unconditional” relationship because it
does not operationally define what that would really mean. A great deal of what
passes for causal modeling suffers from these defects (Preedman 19875 199%; 1997;
1999)
“The Humean approach does even less well with the asymmetrical feature of the
caus relationship because it provides no way to determine asymmetry except tem-
poral precedence. Yet there are many other aspects of the causal relation that seem
‘more fundamental than temporal precedence. Causes not only typically precede their
2 Praise canis do at neces sty INUS conditions becawse an INU factor might only
smtines proce an ef. Ths the shot cic andthe wocden fame ofthe hose might ony
‘Sometimes endo conflagration in which the owe sumed down Inreding potas ces
Would adsl anther jer of compen tour dscsion which woul nly provide more reasons to
doubt the Haren relat approach effects, but they also can be used to explain effects or to manipulate effects while
effects cannot be used to explain causes or to manipulate them.”*
Effects also depend upon causes, but causes do not depend upon effects. Thus, if
a cause does not occur, then the effet will not occur because effects depend on their
‘eauses. The counterfactual, “if the cause did not occur, then the effect would not
‘occur is true. However ifthe effect does not occur, then the cause might stil occur
‘because causes can happen without leading to a specific effect if other features ofthe
situation are not propitious for the effect. The counterfactual, “if the effect did not
‘occur, then the cause would not occur” is not necessarily true. For example, where a
short circuit causes a wooden frame building to burn down, ifthe short circuit does
not occur, then the building will not bura down. But i the building does not burn
isstll possible thatthe short circuit occurred but its capacity for causing fires
‘was neutralized because the building was made of brick. This dependence of effects
‘on causes suggests that an alternative definition of causation might be based upon a
proper understanding of counterfactuals.
down,
5 COUNTERFACTUAL DEFINITION
oF CAUSATION
In a book On the Theory and Method of History published in 1902, Eduard Meyer
claimed that it was an “unanswerable and 50 an idle question” whether the course of
history would have been different if Bismarck, then Chancellor of Prussia, had not
decided to go to war in 3866, By some accounts, the Austro-Prussian-Italian War of
1866 paved the way for German and Italian unification (see Wawro 1996). In reviewing
“Meyer's book in 1906, Max Weber agreed that “from the strict ‘determinist’ point of
view” finding out what would have happened if Bismarck had not gone to war “was
‘impossible’ given the ‘determinants’ which were in fact present” But he went on to
say that “And yet forall tha, is far from being idle’ to raise the question what might
have happened, if, for example, Bismarck had not decided for war. For itis precisely
this question which touches on the decisive element in the historical construction of
reality: the causal significance which is properly attributed to this individual decision
within the totality of infinitely numerous factors (all of which must be just as they
are and not otherwise) if precisely this consequence isto result, and the appropriate
position which the decision is to occupy in the historical account” (Weber 1978,
tit). Weber’ review is an early discussion of the importance of counterfactuals for
understanding history and making causal inferences. He argues forcefully that if
“history isto raise itself above the level ofa mere chronicle of noteworthy events and
® Hausman (99,1 ao catalogs the aspects ofthe asymmetry between cases nd efits
CAUSATION ANDEXPLANATION 233
personalities, it ean only do so by posing just such questions” as the counterfactual in
hich Bismarck did not decide for war."
5. Lewis's Counterfactual Approach to Causation
‘The philosopher David Lewis (19736) has proposed the most elaborately worked out
theory of how causality is related to counterfactuals” His approach requires the
truth of two statements regarding two distinct events X and Y. Lewis starts from the
‘presumption that X and Y have occurred so thatthe “counterfactual” statement,2* "If
X were to occur, then ¥ would occur” is true, The truth ofthis statement is Lewis's
first condition for a causal relationship. Then he considers the truth of a second
counterfactual “If X were not to occur, then Y would not occur either” If this is
true as well, then he says that X causes Y If, for example, Bismarck decided for war
in 1866 and, as some historians argue, German unification followed because of his
decision, then we must ask “If Bismarck had not decided for war, would Germany
have remained divided?” The heart of Lewis's approach isthe set of requirements,
‘described below, that he lays down for the truth ofthis kind of counterfactual.
Lewis's theory has a numberof virtues. It deals directly with singular causal events
and it does not require the examination of a large number of instances of X and
Y. At one point in the philosophical debate about causation, it was believed that
the individual cases such as “the hammer blow caused the glass to break” or “the
assassination of Archduke Ferdinand caused the First World Wat” could not be
analyzed alone because these cases had to be subsumed under a general law (“ham
‘mer blows cause glass to break”) derived from multiple cases plus some particular
facts ofthe situation in order to meet the requirement for a “lawlike” relationship.
The counterfactual approach, however, starts with singular events and proposes that
causation can be established without an appeal to a set of similar events and general
% 1am indebted o Richard Swedberg for pong me awards Weber extraordinary discus,
Lew inde sone ppt for hs theory inthe wrk of Dav Hume fam change ofcourse
na shor pasa in is Emuy Concerning Human Under (8), Hume summed his
gun approach caution by saying that "we may define case tbe an objec. lowed Py
‘ther, and where ll heat sat the sae allowed by jects sino the second and
‘hen be hanged to 3 completly iferemt approach to cxuation by ang “Orin oter words whee
1s chet ad not een the secon had ever exied” 14)- Asan commentators ave note
‘hese mee indeed other words, mpiing an nil lifer notion af etsaton The it poh
une asa wth he constant conjunction of patatv cases and eects cross
‘Sreamitances The second which a counteract! approach, ees pon what would apn in
‘orld where the exo dit ot oct
"Lewis conser statements ie his spat fis theory of counterfactual simply assuming
‘hat sateen inthe subjunctive mood with ue premises and tue concsions see. As noted
‘rir mor theore of counteract have een extend once taterents wih tue pes by
{suming quite reasonably hat they ate tre heir concsion Is tue and ale ethers
"This asp version af Leis theory based upon Lewis 7970 988) aed Hausman
chsh, 234 WENRY E. URADY
Jaws regarding them. The possibilty of analyzing singular causal events is important
for all researchers, but especially for those doing case studies who want to be able to
say somethingabout the consequences of Stalin sueceeding Lenin as head ofthe Soviet
Union or the impact of the butterfly ballot on the 2000 US election.
‘The counterfactual approach also deals directly withthe issue of X's causal “ffi
‘cacy” with respect to ¥ by considering what would happen if X did not occur. The
problem with the theory isthe dificulty of determining the truth or falsity of the
counterfactual “If X were not to occur, then Y would not occur either.” The statement
‘cannot be evaluated in the real world because X actually occurs so thatthe premise
is false, and there is no evidence about what would happen if X did not occu. It
‘only makes sense to evaluate the counterfactual in a world in which the premise is
true. Lewis's approach to this problem is to consider whether the statement is true in
the closest possible world to the actual world where X does not occu. Thus, if X is
a hammer blow and ¥ is a glass breaking, then the closest possible world is one in
hich everything ese is the same except that the hammer blow does not occur. IFin
this world, the glass does not break, then the counterfactual is true, and the harnmer
blow (X) causes the glass to break (Y). The obvious problem with this approach is,
identifying the closest possible world. IF isthe assasination of Archduke Ferdinand:
and Y is the First World War, is it true that the First World War would not have
‘occurred in the closest possible world where the bullet shot by the terrorist Gavrilo
Princip did not hit the Archduke? Or would some other incident have inevitably
precipitated the First World War® And, [Link] to the difficulty, would this “First World
“War” be the same as the one that happened in our world?
Lewis's approach substitutes the riddle of determining the similarity of possible
‘worlds for the neo-Humean’s problem of determining lawlike relationships. To solve
these problems, both approaches must be able to identify similar causes and similar
effects. The Humean approach must identify them across various situations in the
real world. This aspect of the Humean approach is closely related to John Stuart
Mil’s “Method of Concomitant Variation” which he described as follows: “Whatever
phenomenon varies in any manner, whenever another phenomenon varies in some
‘similar manner, i either a cause or an effect ofthat phenomenon, or is connected to
itthrough some fact of causation” (Mill 888,287)" Lewis's theory must also identify,
similar causes and similar effects in the real world in which the cause does occur
and in the many possible worlds in which the cause does not occur. This approach is
® Infact, many authors nw iin that general aatio inoving wie generalizations) can
‘only be understood in terms of singular ausaton:eneral causation isa genealsaton of singular
hatin. poking causes cancer (and onl if) smokes cancers re enaaly caused by their
Smoking” (Males gs 6-7 Se also Sosa and Toke (i) Move generally wheres explanation was
‘once thought italy to supa the need for cul statements many posters now bw that
1 omec ales of casa wl prow s bas fr stable enantio aeSlen 90)
2" The Humean approach abo has ainiss with is Method of Agreement which he described
feo: or more instance of he phenomenon ndrisesigaton have ony one ccumstance
incommon, the cromstanc in which ane al the instances gree the ase (or eet) ofthe ie
Phenomenon (Mil 88,28).
CAUSATION AND EXPLANATION 235
closely related to Mills “Method of Difference” in which: “Ian instance in which the
phenomenon under invest
hhave every circumstance in common save one, that one occurring only inthe former;
the circumstance in which alone the two instances fe, isthe effect, oF the cause, oF
an indispensable part ofthe cause, ofthe phenomenon” (Mill 1888, 280).
Inaddition to identifying similar causes and similar effects, the Humean approach
must determine if the conjunction of these similar causes and effects is accidental
‘or lawlike, This task requires understanding what is happening in each situation and.
‘comparing the similarities and differences across situations. Lewis's approach must,
identify the possible world where the cause does not occur that is most similar to the
real world, This undertaking requires understanding the facts of the real world and.
the laws that are operating init. Consequently, assessing the similarity ofa possible
world to our own world requires understanding the lawlike regularities that govern
four world.” It scems as if Lewis has simply substituted one difficult task, that of
identifying the most similar world for the job of establishing lawfulness.
tion occurs, and an instance in which it does not occur,
5.2 The Virtues of the Counterfactual Definition of Causation
Lewis has substituted one dificult problem for another, but the reformulation ofthe
‘problem has a number of benefits. The counterfactual approach provides new insights,
into what is required to establish causal connection between causes and effects. The
‘counterfactual approach makes it clear that establishing causation does not require
observing the universal conjunction of a cause and an effect. One observation
of a cause followed by an effet is sufficient for establishing causation if it can be
shown that in a most similar world without the cause, the effect does not occur,
“The counterfactual approach proposes that causation can be demonstrated by simply
finding a most similar world in which the absence of the cause leads to the absence of
the effect.
Lewis's theory provides us with a way to think about the causal impact of singular
‘events such as the badly designed butterfly ballot in Palm Beach County, Florida that,
Ted some voters in the 2000 presidential election to complain that they mistakenly
voted for Reform Party candidate Patrick Buchanan when they meant to vote for
Democrat Al Gore. The ballot can be said tobe causally associated with these mistakes
© Mil goes om ono thi the Method of Diference a method of aii experimen” (8)
oti tht for bth Method of Concomitant Variation ad the Method of Diference, Mil
‘enphass the association between case and ee and ot the Mention of which event isthe
{auie and which the effec Mills methods ate diged to detect the syametric aspect of easy but
‘Nelion Gocdman makes ths point in 947 ate on counterfactual and James Fearon 991
s master erpeston of the courteratal appre o research cusses its impleatons for
outer thought experiments in pole ence Alo se Telock and Blin (198).
SIG. H. von Wright notes thatthe counteract ence of causality shows thatthe halla of|
‘ait connection ison nt” (vom Weight 92) 236 HENRY E. RADY
CAUSATION AND EXPLANATION 237
if in the closest possible world in which the butterfly ballot was not used, the vote
for Buchanan was lower than in the real world. Ideally this closest possible world
‘would be a parallel universe in which the same people received a different ballot, but
this, of course is impossible, The next-best thing i a situation where similar people
employed a different ballot. Infact, the butterfly ballot was only used for election day
voters in Palm Beach County. It was not used by absentee voters. Consequently, the
results for the absentee voting can be considered a surrogate forthe closest possible
world in which the butterfly ballot was not used, and in this absentee voting world,
voting for Buchanan was dramatically lower, suggesting that atleast 2000 people who
preferred Gore—more than enough to give the election to Gore—mistakenly voted.
for Buchanan on the butterfly ballot.
‘The difficult question, of cours, is whether the absentee voting world can be con-
sidered a good enouigh surrogate forthe closest possible world in which the butterfly
ballot was not used." The counterfactual approach does not provide us with a clear
sense of how to make that judgment.” But the framework does suggest that we should
consider the similarity of the election day world and the absentee voter world. To do
this, we can ask whether election day voters are different in some significant ways
from absentee voters, and this question can be answered by considering information
‘on their characteristics and experiences. In summary, the counterfactual perspective
allows for analyzing causation in singular instances, and it emphasizes comparison,
‘hich seems difficult but possible, rather than the recondite and apparently fruitless
investigation of the lawfulness of statements such as “All ballots that place candidate
‘names and punch-holes in confusing arrangements wil lead to mistakes in casting
5.3 Controlled Experiments and Closest Possible Worlds
“The difficulties with the counterfactual definition are identifying the characteristics
‘of the closest possible world in which the putative cause does not occur and finding
an empirical surrogate for this world. For the butterly ballot, sheer luck led a team of
researchers to discover that the absentee ballot did not have the problematic features
‘of the buterfly ballot But how can we find surrogates in other circumstances?
© Foam argument thatthe absent vote are an excellent srr se Wand ta 9
° Ins bock om counteracts Lewis ony clas tht silty dente posi but he
oes na prvi yume on how to mak them He amie ht hie tion gue bt he ans
ieisno tanderstood et compaative snarky isnt l-andestol is ages gen 3
swellndersood way. Therefore iss the Sort of primitive hat we must seo lve corte analy
‘tsomethng that se nent age Lew ro -Un er work Lets (9791988)
Formutes ome rales fo simi jadgments, but thy do ot seem very welt and thes
(Bennet 18.
For the sory of how the dieencs between the election day and ante blot were discovered
see Brady a (at),
(One answer is controlled experiments. Experimenters can create mini-losest-
possible worlds by finding two or more situations and assigning putative causes
(called “treatments”) to some situations but not to others (which get the “control”). If
in those cases where the cause C occurs, the elect E occurs, then the fist requirement
‘of the counterfactual definition is met: When C occurs, then E occurs. Now, if the
situations which receive the control are not different in any significant ways from
those that get the treatment, then they can be considered surrogates for the closest
‘possible world in which the cause does not occur. [Fin these situations where the cause
does not occur, the effet F does not occur either, then the second requirement of
the counterfactual definition is confirmed: Inthe closest possible world where C does
not occur, then does not occur. The crucial part ofthis argument is that the control
situation, in which the cause does not occur, must be a good surrogate forthe closest
possible world tothe treatment,
‘Two experimental methods have been devised for ensuring closeness between the
treatment and control situations, One is classical experimentation i which as many
circumstances as possible ate physically controlled so that the only significant difer-
ence between the treatment and the contro isthe cause. Ina chemical experiment,
for example, one beaker holds two chemicals and a substance that might be a catalyst
and another beaker of the same type, in the same location, atthe same temperature,
and so forth contains just the two chemicals in the same proportions without the
suspected catalyst. Ifthe reaction occurs only in the first beaker, itis attributed to
the catalyst. The second method is random assignment of treatments to situations
so that there are no reasons to suspect that the entities that get the treatment are
any different, on average, from those that do not. We discuss this approach in detail
below.
5.4 Problems with the Counterfactual Definition™
Although the counterfactual definition of causation leads to substantial insights about
causation, it also leads to two significant problems. Using the counterfactual dfini-
tion as it has been described so far the direction of causation cannot be established,
and two effects of common cause can be mistaken for cause and effect. Consider,
{or example, an experiment as described above. In that case, inthe treatment group,
when C occurs, E occurs, and when E occurs, C occurs. Similarly, in the control
‘group, when C does not occur, then E does not accur, and when F does not occur,
then C does not occur. Infact, there is perfect observational symmetry between cause
and effect which means that the counterfactual definition of causation as described so
far implies that C causes E and that E causes C. The same problem arises with to
‘effects of a common cause because ofthe perfect symmetry in the situation. Consider,
for example, arse in the mercury ina barometer and thunderstorms. Each isan effect
> Tinsecton rei esi pon Hausman (98 especialy hs 47] and Lewis (ur), 238 HENRY E. BRADY
of high pressure systems, but the counterfactual definition would consider them to be
«causes of one another.”
“These problems bedevil Humean and counterfactual approaches. [Fwe accept these
approaches in thei simplest forms, we must lie witha seriously incomplete theory
‘of causation that cannot distinguish causes from effects and that cannot distinguish
two effects of a common cause from real cause and effect. That is, although the
counterfactual approach can tell whether two factors A and B are causally con-
nected in some vray it cannot tell whether A causes B, B causes A, or A and.
Bare the effects of a common cause (sometimes called spurious correlation). The
reason for this is thatthe truth of the two counterfactual conditions described so
far amounts to a particular pattern of the crosstabulation of the two factors A and.
B. In the simplest case where the columns are the absence or presence ofthe first
factor (A) and the rows are the absence or the presence of the second factor (B),
then the same diagonal pattern is observed for situations where A causes B or B
‘eauses A, of for A and B being the effects of a common cause, In all three cases,
‘we either observe the presence of both factors or their absence. It is impossible
from this kind of symmetrical information, which amounts to correlational data,
to detect causal asymmetry or spurious correlation. The counterfactual approach as,
clucidated s0 far ike the Humean regularity approach, only describes a necessary
condition, the existence ofa causal connection between A and B, for us to say that A.
causes B.
Requiring temporal precedence can solve the problem of causal direction by simply
choosing the phenomenon that occurs first as the cause, but it cannot solve the
problem of common cause because it would lead to the ridiculous conclusion that
since the mercury rises in barometers before storms, this upward movement in the
‘mercury must cause thunderstorms. For this and other reasons, David Lewis rejects,
using temporal precedence to determine the direction of causality Instead, he claims
that when C causes E but not the reverse “then it should be possible to claim the
falsity of the counterfactual ‘If E did not occur, then C would not occur!” This
‘counterfactual is different from “if C occurs then F occurs” and from “if C does not
‘occur then E does not occue” which, as we have already mentioned, Lewis believes
must both be true when C causes E. The required falsity of “IE did not occur, then.
would not occur” adds a third condition for causality. This condition amounts
to finding situations in which C occurs but F does not—typically because there is,
some other condition that must occur for C to produce E. Rather than explore this
strategy, we describe a much better way of establishing causal priority in the next
Th itharometicpressaress thunderstorms ocur and vie vers. Furthermore, barometric
pressure doesnot rise then thunders do not ocr and vce ers, Ths, bythe counieractal
Uentonseah isthe cao the oe, To lif ters we have ignored heft hat tere nat
‘evfectdeerministiceltonship between high pressure systems and thunderstorms)
‘os impl ty thi paragraph theres cas connection beeen A and B when either A cases
2B causes sor Sad athe elects of common ets, Set Hasan 998, 5-6)
29
6 EXPERIMENTATION AND THE
MANIPULATION APPROACH TO CAUSATION
In an experiment, there is a readily available piece of information that we have
‘overlooked so far because its not mentioned inthe counterfactual approach The
factor that has been manipulated can determine the direction of causality and help to
ruleout spurious correlation. The manipulated fctor mustbe the cause. Isard to
cxaggerate the importance of ths insight. Although philosophers are uncomfortable
‘ith manipulation and agency approaches to causality because they put people (asthe
manipulators) tthe center of our understanding of causality, there can be itle doubt
about the power of manipulation for determining causality. Agency and manipu
lation appreaches to causation (Gaskng 1955 von Wright 1974; Menzies and Price
1993) elevate this insight into ther definition of causation, For Gasking “the notion
‘of causation is essentially connected with our manipulative techniques for producing
results (935, 43), and for Menzies and Price “events are causally related just in case
the situation involving them possesses intrinsic features that either support a means
‘end relation between the events sis, rae identical with (or closely similar to) those
‘of another situation involving an analogous pair of means-end related evens” (1993
197). These approaches focus on establishing the direction of causation, but Gasking’s
‘metaphor of causation as “recipes” also suggests an approach towards establishing the
symmetric, regularity aspect of eausaton. Causation exists when theresa recipe that
regularly produce effects fom causes.
Perhaps our ontological definitions of causality should not employ the concept af
agency because most ofthe causes and effects in the universe go their merry way
without human intervention, and even our epistemological methods often discover
