Episteme, 14, 1 (2017) 103–127 © Cambridge University Press

doi:10.1017/epi.2015.53

testifying understanding
kenneth boyd
[Link]@[Link]

abstract
While it is widely acknowledged that knowledge can be acquired via testimony, it
has been argued that understanding cannot. While there is no consensus about
what the epistemic relationship of understanding consists in, I argue here that
regardless of how understanding is conceived there are kinds of understanding
that can be acquired through testimony: easy understanding (e.g. understanding sim-
ple and mundane information) and easys understanding (e.g. understanding infor-
mation that might be complex, but is nevertheless easy for an expert S). I address
a number of aspects of understanding that might stand in the way of being able
to acquire understanding through testimony, focusing on understanding’s paradig-
matic form and what it means to say that in order to understand something you
need to “grasp” some information or the relationship between bits of information.
I argue that in cases of both easy and easys understanding, no aspect of understand-
ing stands in the way of it being able to acquire it through testimony. As a result,
while not all understanding can be acquired through testimony in all instances
and for all subjects, this failure of acquisition is only a product of the complexity
of the relevant information or one’s unfamiliarity with it, and not a product of
the epistemic relationship of understanding.

1. directness and indirectness

Let’s say that you and I are watching baseball, a sport about which you know only a little
but I know a lot. With a runner on third base and two outs, the pitcher intentionally
throws four balls in a row, advancing the batter to rst base. You are confused: why,
you ask, would the pitcher give his opponent a free trip to rst base? I explain to you
that the batter is an extremely good hitter, and so it is safer to put him on rst base
and remove the possibility of him driving in a run, especially since there is only one out
remaining in the inning, and because the next batter is a relatively poor hitter. As a result
of our conversation, you come to know some new things: you have gained, as a result of
my testimony, some new knowledge about baseball.
Say now that I not only know, but understand a lot about baseball, as well: perhaps I
have watched a lot of baseball, maybe even played a little, am familiar with the rules, strat-
egies, and roles of the relevant players, etc. Let us also say, then, that I not only know why
the pitcher intentionally walked the batter, I also understand why he did. Although it
seems perfectly natural to say that I have this understanding, it’s not terribly clear what
this understanding consists in. It seems, however, that the following things about

[Link] Published online by Cambridge University Press

e p i s t e m e v o l u m e 14–1 103
 kenneth boyd

understanding are true. First, it seems that my background familiarity with and knowledge
about baseball are required for me to have my understanding: someone who knows
almost nothing about baseball would not have the understanding that I do. Second, my
understanding seems to consist in more than just a collection of things that I know: under-
standing why the pitcher intentionally walked the batter seems to require, to use a com-
mon metaphor, that I be able to grasp connections between relevant bits of knowledge
that I have, say the knowledge I have of the rules of baseball, baseball strategies, and
the relative strengths and weakness of the players.1 Finally, it seems that while I under-
stand a lot about baseball, it also seems that in general I understand less than I know,
as understanding seems to be harder to get than knowledge.
Now, while it seems that you can acquire various bits of knowledge about baseball
through my testimony, it seems much less plausible to say that you can acquire under-
standing in the same way. The best, it seems, that you can do is to acquire as much
knowledge as possible from me, and on that basis you can potentially come to under-
stand why the pitcher did what he did. But it’s still up to you to acquire this under-
standing for yourself: you have to put together the pieces, a process that perhaps
involves other things I’ve told you, or other knowledge that you have already.
Again, understanding seems to require that you grasp some fact or information, or
relationships between other facts or other information, and this is not something I
can do for you. This is not something that you have to do when acquiring knowledge:
while testimony can be a direct source of knowledge, it seems that it cannot be a direct
source of understanding.2
Recently, those writing on understanding and testimony have adopted just this view of
the relationship between them. Consider, for example, the following from Linda
Zagzebski (2008):

Knowledge can be acquired by testimony, whereas understanding cannot be. A conscientious

believer can obtain a true belief on the testimony of another, and given the right conditions,
can thereby acquire knowledge. . .Understanding cannot be transmitted in that way. In fact, under-
standing cannot be given to another person at all except in the indirect sense that a good teacher
can sometimes recreate the conditions that produce understanding in hopes that the student will
acquire it also. (145–6)

1 Talk of “grasping” is ubiquitous in discussions of understanding. See, for example, Kitcher (1989),
Kvanvig (2003), de Regt and Dieks (2005), Grimm (2006, Forthcoming), Elgin (2007), Zagzebski
(2008), Hills (2009, Forthcoming), Khalifa (2013), amongst others. I will look at grasping in much
more detail in section 2.5.
2 The question of whether something is a direct or indirect source of an epistemic relationship is different
from a related question in epistemology, namely that concerning whether testimony is a basic source of
knowledge. Those who argue that testimony is a basic source of knowledge fall into the non-
reductionist camp, and include testimony along with sources of knowledge like perception and memory.
Those who argue that it is not are reductionists, roughly those who think that the justication we get for
our beliefs acquired through testimony is reducible to the justication provided by something else (for
an overview of the debates between reductionists and anti-reductionists, see Lackey (2006)). I do not
take a stance on the reductionism/non-reductionist debate here (although I do return to it later,
briey). Regardless, both reductionists and non-reductionists would agree that testimony is a direct
source of knowledge.

104 e p i s t e m e v o l u m e 14–1
[Link] Published online by Cambridge University Press
 testifying understanding

Allison Hills (2009) argues in the same way:

If you are attempting to gain knowledge, testimony can serve as the justication for your own
belief but it is not usually a good way of acquiring . . . understanding. Understanding why p
will not – cannot – have the same relationship to testimony as knowing why p. (19–20)

Duncan Pritchard (Pritchard et al. 2010), although not arguing explicitly for the view that
understanding cannot be acquired through testimony, presents a case that exemplies the
worries of Hills and Zagzebski:

Suppose that I understand why my house burned down, know why it burned down, and also
know that it burned down because of faulty wiring. Imagine further that my young son asks
me why his house burned down and I tell him. He has no conception of how faulty wiring
might cause a re, so we could hardly imagine that merely knowing this much sufces to afford
him understanding of why his house burned down. (81)

Here, then, is what seems to be a plausible view:

Indirectness: Testimony cannot be a direct source of understanding; at best, it can be an indirect

source of understanding by laying the groundwork for potential understanding.

Despite its prima facie plausibility, I’m going to argue that Indirectness is, in general, false:
there is reason to think that at least some kinds of understanding (and not just the grounds
for understanding) can be acquired directly through testimony.
An obstacle that impedes a thorough investigation of the relationship between under-
standing and testimony is that there is no consensus about what constitutes the epistemic
relationship of understanding. Nevertheless, I will argue that no matter how understand-
ing is construed, there are two general types of understanding that can be acquired
through testimony: easy understanding and easys understanding. Easy understanding is
that which is concerned with basic facts or concepts and can be acquired without requir-
ing any specialized background knowledge or cognitive abilities;3 while easys understand-
ing is understanding that is easy for a particular subject S, given their background
familiarity with the relevant subject matter, knowledge, skills, and abilities (for example,
something might be easys for an expert that would not be easy for a layperson). I argue
that if these kinds of understanding can be acquired through testimony, then the reason
why Indirectness seems to be true is that understanding is often presented as something
that is difcult, and things that are difcult to understand require substantial work on
the part of the person doing the understanding. I conclude by considering the implications
of my arguments on views of understanding that portray it as a signicant cognitive
achievement and, by extension, a uniquely valuable epistemic concept.
To discuss the relationship between understanding and testimony we need to discuss
both understanding and testimony. Here I will be concerned almost exclusively with the
former. In terms of the latter, I will assume for the sake of discussion that in order for

3 Zagzebski (2008) mentions in passing that there may very well be kinds of “easy understanding”: her
example is of “understanding a stop sign in the United States” (144). However, we differ insomuch as
Zagzebski does not think that even easy understanding can be transmitted through testimony.

[Link] Published online by Cambridge University Press

e p i s t e m e v o l u m e 14–1 105
 kenneth boyd

a subject to acquire an epistemic relationship with a proposition or propositions via testi-

mony requires that the speaker possess that epistemic relationship with the relevant propos-
ition(s), that the hearer conceives of the speaker in some way as trustworthy (either by having
positive reasons to believe that the speaker is trustworthy, or by lacking any reasons to doubt
that she is), and that the hearer also possesses the necessary background concepts or beliefs
to properly interpret what the speaker is saying. The purpose of this bare-bones conception
of testimony is merely to show that a conception of testimony that permits the acquisition of
knowledge is also one that permits the acquisition of understanding.4
Philosophical work on the epistemic relationship of understanding is relatively nascent;
as a result, there are some fundamental aspects of understanding that are a matter of sign-
icant debate. I will consider ve such aspects here: understanding’s relationship to
knowledge; whether understanding is factive; the extent to which understanding is suscep-
tible to epistemic luck; understanding’s paradigmatic form; and understanding’s require-
ment of a grasping relation. I will consider each aspect in turn as it relates to the
possibility of acquiring understanding through testimony. In other words, for each of
these aspects of understanding, the relevant question becomes: does this aspect, in general,
stand in the way of being able to acquire understanding directly through testimony? I will
argue that for each of these questions the answer is “no.”

2. five open questions about understanding

2.1 Is understanding a kind of knowledge?
The rst major question concerns understanding’s relationship to knowledge – speci-
cally, whether understanding is a species of knowledge. On one view, to understand some-
thing is just to know more about it: one might have to know the reasons why it’s true, or
know how it relates to other aspects of the relevant domain, etc. (see, for example, Lipton
2004; Grimm 2006; Kelp Forthcoming).5 For example, considering the baseball case again,
understanding why the pitcher intentionally walked the batter might consist in knowing all
of the following: that ‘the batter is a very good hitter,’ that ‘throwing four balls on purpose is
an intentional walk,’ as well, perhaps, as the relationship between this and other knowledge,
say ‘the pitcher threw four balls intentionally because the batter is a very good hitter.’ Here,
then, is a very simple argument for the possibility of acquiring understanding through
testimony: if understanding is a species of knowledge, and knowledge can be acquired
through testimony, then at least some understanding can be acquired through testimony.
There are ways in which this argument could be invalid. We might think, for instance,
that while some knowledge can be acquired through testimony, not all of it can: for
example, some have argued that moral knowledge cannot be acquired through testimony

4 Jennifer Lackey (2008) has argued (in my mind, convincingly) that the metaphor of “transmitting”
knowledge leads to faulty conceptions of testimony: on Lackey’s conception, knowledge acquired
through testimony comes from the words themselves, not the epistemic relationship possessed by the
speaker of those words. To leave room for this view, I discuss the process of “acquiring through testi-
mony” rather than “transmitting through testimony.” Lackey’s view is too nuanced to address in detail
here; nevertheless, if Lackey is correct, then I see no reason to think that we cannot gain understanding
through words, as well.
5 I will return to these views in more detail in what follows.

106 e p i s t e m e v o l u m e 14–1
[Link] Published online by Cambridge University Press
 testifying understanding

(see Hopkins 2007)6. Understanding, then, might simply turn out to be a kind of knowl-
edge that cannot be acquired through testimony. If this happens to be the case, then it
must be because of some aspect of understanding that distinguishes it from other kinds
of knowledge that can be so acquired. Answering the question of whether understanding
is a kind of knowledge, then, will not by itself shed any light on the question of whether
understanding can be acquired through testimony, as we would still need to gure out
what aspect of understanding stands in the way of being able to acquire it in this way.
We need to look at other aspects of understanding instead.7

2.2 Is understanding factive?

Knowledge is factive. Some, however, have argued the understanding is not. Linda
Zagzebski (2008) is perhaps the most prominent proponent of this characterization of
understanding: she argues that while understanding requires grasping the relevant connec-
tions between various pieces of information, as well as the exercising of certain kinds of
skills, neither of these requirements entail truth. Catherine Elgin (2009) also expresses
reservations about understanding’s factivity, since understanding is not clearly propos-
itional (an issue I will return to below). By itself, a difference in factivity would be enough
to distinguish understanding from knowledge as a unique epistemic relationship. The fac-
tivity of an epistemic relationship, however, does not seem, by itself, to be capable of pre-
venting that relationship from being acquired through testimony. For example, knowledge
is factive and can be acquired through testimony, whereas belief is not factive but can also
be acquired through testimony. So the question of factivity is not of primary importance
with regard to the question at hand.

2.3 Is understanding susceptible to luck?

It is widely accepted that knowledge is incompatible with at least some kind of luck. Two
kinds of luck that have been proposed as incompatible with knowledge are Gettier-style
luck and environmental luck. The former is exemplied by Edmund Gettier’s (1963): in
general, Gettier-style luck occurs when one has a true belief, but one’s belief is not true
because of the facts that make it true. Environmental luck, on the other hand, typically
involves cases in which one’s belief is true, and is true because of the facts that make it
true, but in which the belief could have easily turned out to be false.8 These are merely
rough specications; regardless, while knowledge is typically taken to be incompatible
with both Gettier-style luck and environmental luck, various authors have argued that

6 Hills (2009), however, argues that the view that moral knowledge cannot be transmitted through tes-
timony either reduces to skepticism about the existence of moral knowledge or an implausible view
about testimony. I will not take a stance on this debate here.
7 I do not take the open questions I address here to be necessarily exhaustive of relevant questions per-
taining to the nature of understanding. Rather, I consider those aspects of understanding that are
most frequently discussed by those writing on the topic, as well as the aspects that seem most likely
to stand in the way of acquiring understanding through testimony.
8 The paradigmatic environmental luck case is the barn façade case: in a case in which one sees a barn,
forms the true belief that ‘there is a barn,’ but, unbeknownst to the belief-former, is looking at the only
genuine barn amongst a sea of barn façades, is one in which the subject does not possess knowledge (see
Goldman 1976).

[Link] Published online by Cambridge University Press

e p i s t e m e v o l u m e 14–1 107
 kenneth boyd

one can have understanding even if one acquired it in one or both of these lucky ways. For
example, Kvanvig (2003) and Morris (2012) argue that understanding is compatible both
with Gettier-style and environmental luck (that is to say, one could have understanding of
some information even if one’s understanding is true but not because of the reasons that
make it true, and even if that information could have easily turned out to be false), where-
as Grimm (2006) and Pritchard et al. (2010) argue that understanding is incompatible
with Gettier-style luck, but compatible with environmental luck.
As is the case with factivity, susceptibility to either Gettier-style or environmental luck
does not seem, on its own, to be able to determine whether one can acquire understanding
via testimony. If anything, it seems that the less susceptible a relationship is to being
undermined by luck, the easier it becomes for one to acquire it. If knowledge is incompat-
ible with both Gettier-style and environmental luck, and knowledge can be acquired
through testimony, then it is hard to see how an epistemic relationship being susceptible
to only one or neither of these kinds of luck could prevent it from being acquired through
testimony.
While differences between understanding and knowledge in terms of factivity or luck
can be dismissed quickly, the other potential differences require signicantly more care.

2.4 What is the object of understanding?

Traditionally, when epistemologists talk about knowledge they talk about propositional
knowledge – or “knowledge-that” – and the paradigmatic form of a knowledge attribu-
tion has been taken to be of the form “S knows that p”.9 There is much less consensus
when it comes to the object and paradigmatic form of understanding. The three most
popular candidates are illustrated in the following examples:

Understanding-why: With a runner on third base and two outs, a very good hitter comes to the
plate. As a spectator in the stands (and long-time baseball fan), you see that the pitcher, not want-
ing to risk giving up a run, intentionally throws four balls in a row to give the batter a free trip to
rst base. Given your background familiarity with and knowledge of baseball, it seems natural to
say that “You understand-why the pitcher intentionally walked the batter.”

Understanding-how: With a runner on third base and two outs, a very good hitter comes to the
plate. As the pitcher, you don’t want to give up another run, so you gure it’s safer to give the
batter a free trip to rst base than to give him the opportunity to drive in a run. In your experience,
the best way to do this is to signal to the catcher, and then throw four pitches well outside of the
strike zone, making sure that the batter has no chance of hitting them. Given your background
familiarity with and knowledge of baseball, it seems natural to say that “You understand-how
to pitch strategically.”

9 This is not the only kind of knowledge attribution we can make: for instance, we often make knowledge
attributions that someone knows how to do something, or that one knows-wh something, i.e. that one
knows where something is, when something is, what something is, or who someone is. It has been
argued that knowledge-wh is propositional (for an overview of this debate, see Brogaard 2009), and
it is a matter of signicant debate whether knowledge-how and knowledge-that are reducible to one
another or are distinct (see, for instance, Stanley and Williamson 2001). I will discuss these issues
briey in what follows, but I won’t take a stand on any of the debates.

108 e p i s t e m e v o l u m e 14–1
[Link] Published online by Cambridge University Press
 testifying understanding

Understanding-□ (where “□” is some object, concept, or phenomenon): With a runner on third
base and two outs, a very good hitter comes to the plate. As the coach of the team on defense,
you can see that the next batter is a relatively poor hitter, so you signal to the pitcher that he
should intentionally walk the batter. Given your background familiarity with and knowledge of
baseball, it seems natural to say that “You understand-□ baseball” (or, perhaps more naturally
still, that “You understand baseball”).10

Various authors writing on understanding have taken different forms of understanding

to be paradigmatic: for example, Pritchard et al. (2010) and Hills (Forthcoming) take
understanding-why to be the paradigmatic form of understanding; Zagzebski (2008)
takes understanding-how to be paradigmatic; and Kvanvig (2003), Wilkenfeld (2013),
and Kelp (Forthcoming) take understanding-□ to be paradigmatic. While all of these
forms of understanding seem natural in the relevant context, it is up for debate whether
one of these forms is truly paradigmatic, whether some forms are derivative of or reducible
to others, or whether “understanding” is polysemous. Without taking a stance on which,
if any, of these forms is truly paradigmatic, the question becomes whether any of these
forms would entail that the relevant form of understanding cannot be acquired through
testimony.
It seems plausible that they might. Here’s the problem. Testimony, as it is typically con-
ceived, is concerned with propositions: testimony is conveyed through uttered sentences
that express propositions, so it might be thought that whatever I acquire via testimony
must itself be something that is propositional. Knowledge-that ts the bill: if you know
that p, tell me that p, and the right background conditions are in place, I can come to
know that p as a result of your testimony. Is understanding concerned with propositions?
It’s not clear. Understanding certainly has to be about something, be it some bits of infor-
mation or subject matter, and whatever one’s understanding happens to be about could
potentially be either a set of propositions or something expressible in terms of proposi-
tions. There is, of course, the requirement that one also grasp the relevant connections
between pieces of information or the subject matter as a whole in order to have the rele-
vant understanding. Thus there seem to be two main components of an instance of under-
standing: an informational component and a grasping component. While the grasping
component is the focus on the next section, in this section I will argue for the following:
regardless of what the paradigmatic form of understanding happens to be, the informa-
tional component of an instance of understanding is something that can be acquired
through testimony. If there is something that stands in the way of understanding being
acquirable through testimony, then, it is not a consequence of its paradigmatic form.
Consider rst the case of understanding-why: Pritchard (Pritchard et al. 2010) states
that “knowledge . . . is concerned with propositions, whereas understanding usually

10 The kinds of things that could be included in the place of the “□” are multitudinous and disparate.
Kvanvig (2003), for instance, gives the examples of “understanding quantum mechanics” and “under-
standing my partner” (we will also see many more instances of understanding-□ in what follows). The
important aspect of the understanding-□ form is that understanding is supposed to be a relationship
that one has towards a thing, or one’s representation of a thing, concept, or phenomenon. I address
this in more detail below. However, it is an open question as to whether the kinds of things that
can take the place of the □ can all be spoken of in a unied way: Grimm (Forthcoming), for example,
argues that there is reason to think that there are differences between understanding people and under-
standing scientic theories. I will not engage with this debate here.

[Link] Published online by Cambridge University Press

e p i s t e m e v o l u m e 14–1 109
 kenneth boyd

isn’t, at least not directly anyway” (74). Pritchard is saying that while knowledge is para-
digmatically taken to be concerned directly with propositions in the form of knowledge-
that attributions, “it is rare to talk of understanding that p” (74). These considerations by
themselves, however, do not necessarily speak against the propositionality of the informa-
tional content of understanding-why: Pritchard’s paradigmatic form of an understanding
attribution is of the form “understanding-why X is the case”, and it’s plausible that to
understand-why X is the case one’s understanding is directed towards a set of proposi-
tions, namely those reasons as to why X is the case. For example, my understanding-why
the pitcher intentionally walked the batter might have as its object a set of propositions
including ‘there are two outs,’ ‘the batter is a very good hitter,’ ‘the next batter is a
weak hitter,’ etc., along with relations between them, perhaps ‘the pitcher threw four
balls in a row because there are two outs . . .’. Indeed, without having some kind of rela-
tionship with these propositions it is difcult to see how anyone could be said to under-
stand why the pitcher intentionally walked the batter.
Of course, all of this is in line with Indirectness: if one holds the view that all testimony
can do is to provide the basis of understanding, then one would presumably be happy with
saying that the relevant information one has understanding of can be acquired through
testimony. What’s important for the Indirectness proponent is that there is an aspect of
understanding that cannot be acquired via testimony. I examine this argument in detail
in section 3; for now, however, it is enough to note that if understanding-why is the para-
digmatic form of understanding, then the fact that the object of understanding is some-
thing of the form “X being the case” does not stand in the way of understanding being
acquirable directly through testimony.
What about the other forms? Is the object of understanding-how propositional? It’s not
clear. Consider an analogous question, that of the object of knowledge-how: some have
argued that knowledge-how is propositional because it is simply a kind of knowledge-that
(Stanley and Williamson 2001, for instance, argue that knowledge-how is just knowledge-
that conceived of under a “practical mode of presentation”), whereas others have argued
that while knowledge-how is not necessarily reducible to knowledge-that, it is nevertheless
still propositional (Hawley 2010, for example, considers this possibility). If the same could
be said for understanding-how – i.e., that understanding-how is either reducible to a form
of propositional understanding or is still expressible in some propositional way – then
again there does not seem to be a barrier, in terms of its propositionality, to it being
acquirable through testimony. However, there is also a sense in which knowledge-how
is concerned with a skill or an ability; analogously, Zagzebski, who seems to take
understanding-how to be the paradigmatic form of understanding, argues that under-
standing often involves the mastery of a skill or technê (2008: 144). Skills, abilities, and
techni, however, might not be the kinds of things that are propositional or reducible to
something that is. If this is the case, then understanding-how might not be able to be
acquired through testimony because it is non-propositional.
It does not seem likely that one can gain a mastery of a skill that one was previously
unfamiliar with through a single testimonial act, and in this sense it seems true that one
cannot acquire understanding directly through testimony. However, even if this under-
standing necessarily involves the possession of a skill or ability, there is still reason to
think that one can acquire new understanding-how by acquiring new information, since
once one already possesses a set of skills within a relevant domain one can come to under-
stand how to do new things within that domain by applying one’s skills to new

110 e p i s t e m e v o l u m e 14–1
[Link] Published online by Cambridge University Press
 testifying understanding

information. For example, the baseball pitcher who understands how to throw a curveball
can come to understand how to throw a slider by applying his baseball skills to new infor-
mation; the boxer who understands how to beat a right-handed opponent can come to
understand how to beat a southpaw by applying her boxing skills to new information;
the math student who understands how to factor a binomial can come to understand
how to factor a trinomial by applying their mathematics skills to new information; etc.
I will return to this idea below in sections 3 and 4; for the time being, the important
point is that even if the paradigmatic form of understanding is understanding-how,
since understanding-how can plausibly be acquired by applying already-possessed skills
to new information (one does not always have to develop new skills when gaining new
understanding-how), there is nothing about the object of understanding-how that stands
in the way of it being acquirable through testimony.11
The last form of understanding is also the one that seems to be the most clear-cut in
terms of propositionality: the case of understanding-□. Above, I gave the example of
understanding baseball, where the object of understanding was the game of baseball itself.
The game of baseball, however, is straightforwardly not a proposition, and thus it seems
equally straightforward that an understanding of baseball cannot be acquired through tes-
timony. We could, however, approach understanding-□ in the same way that we
approached the other forms of understanding, namely to consider whether
understanding-□ is either expressible in terms of propositions or is reducible to a kind
of propositional understanding: for example, understanding baseball might just consist
in a collection of instances of understanding-why or understanding-how (e.g. I understand
how to win games, I understand why my opponents do what they do in certain situations,
etc.); alternatively, when baseball is the object of my understanding, I might consider the
game as represented by a set of rules and strategies, which is something that, presumably,
can be expressed propositionally.
Several authors who take understanding-□ to be paradigmatic explicitly argue that
understanding-□ is propositional, or can be expressed in terms of something proposition-
al. Wilkenfeld (2013), for example, argues that the correct form of an attribution of
understanding is “understanding x,” where potential candidates for x include such dispar-
ate objects as formal theories, proofs, artistic objects, narratives, historically situated
events, artifacts, individual people, games, physical theories, actions, the meaning of a
word, and a language (998). Nevertheless, Wilkenfeld argues that understanding is a
“mental phenomenon,” since the proper object of understanding is not the items them-
selves in the aforementioned list, but rather one’s representation of those items. Since men-
tal representations are at least sometimes propositional, then Wilkenfeld’s version of
understanding-□ is one way of conceiving understanding-□ as propositional.12

11 Again, this is all still in line with the Indirectness supporter: said supporter might agree with everything
said up until this point, because acquiring understanding still involves active cognitive work involved
in grasping the information one has received. I will address this in section 3. For now, the point is sim-
ply that if understanding-how, in virtue of its object, does not stand in the way of acquiring it through
testimony.
12 One might object that not all mental representations are propositional, and thus even if the object
of understanding-□ is a mental representation of a phenomenon, this does not in itself establish that
understanding-□ is propositional. Wilkenfeld, however, clearly thinks that at least some cases of
understanding-□ are propositional, as he argues that propositions or sets of propositions themselves
can be included in the list of things that are the potential object of understanding-□ (2013: 1000).

[Link] Published online by Cambridge University Press

e p i s t e m e v o l u m e 14–1 111
 kenneth boyd

Similarly, Kelp (Forthcoming) presents a version of understanding-□ in which the

object of understanding is a phenomenon. Again, the list of possible phenomena are
diverse: persons, objects, events, processes, and instantiations of properties and relations
all make the list (10). Regardless, Kelp maintains that “for every phenomenon, no matter
its metaphysical nature, there is a set of true propositions that describe it” as well as “a set
of true propositions describing its place in a broader nexus of phenomena,” where the for-
mer set of propositions constitutes a phenomenon’s description, and the latter set consti-
tutes its story (11). For Kelp, understanding a phenomenon is expressed in terms of its
“full account,” which is the union of its description and story. Again, without evaluating
the specic merits of Kelp’s account, we can see that it provides another propositional con-
ception of understanding-□.
Nevertheless, one might worry that there are certain objects or phenomena that cannot,
for whatever reason, be plausibly expressed in terms of a set of propositions, perhaps
because the objects are particularly complex. As we will see below, I leave open the pos-
sibility that understanding a very complex object or phenomenon can be very difcult, and
that difcult understanding, in any of its forms, might not be able to be acquired through
testimony. But understanding-□ need not involve particularly complex objects, and might
not be particularly difcult to obtain. We can understand-□ simple objects, things, or phe-
nomena, and, as was the case with the other forms of understanding, while
understanding-□ does not seem to be concerned with individual propositions, there are
still individual, discrete instances of understanding-□. For example, the baseball coach
can understand player management, the batting coach can understand batting, and the
pitching coach can understand pitching, with none of them understanding baseball in
its entirety; one physics student can understand the Schrodinger’s cat thought experiment,
another can understand wave functions, and another can understand wave-particle dual-
ity, with none of them understanding quantum mechanics in its entirety; etc. As was the
case with understanding-how, while it seems that an understanding of the entirety of base-
ball cannot be acquired through a single testimonial exchange, with the right kind of back-
ground knowledge and abilities in place one could potentially acquire new instances of
understanding-□ by applying one’s background knowledge and abilities to new pieces
of information.
Kvanvig (2003), for instance, argues that understanding-□ is concerned with “the ways
in which pieces of information are connected with each other” (3). Since one can gain new
information through testimony, then it seems that one could at least gain the requisite
material for new understanding-□ through a testimonial exchange, even if the □ that
one is understanding is not a proposition or set of propositions. Other conceptions of
understanding-□ also seem to permit the acquisition of new understanding by acquiring
new information: gaining new information could result in the acquisition or modication
of a mental representation if we adopt Wilkenfeld’s model, and could create a story or
description, or modify an existing story or description of a phenomenon if we adopt
Kelp’s model. That an instance of understanding-□ should have an informational compo-
nent, then, means that if there is anything standing in the way of being able to acquire
understanding-□ through testimony it is not a consequence of the paradigmatic form of
understanding.
I have argued in this section that regardless of what the paradigmatic form of under-
standing happens to be, there is no barrier to acquiring the informational component of
an instance of understanding through testimony. However, I have also argued that it is

112 e p i s t e m e v o l u m e 14–1
[Link] Published online by Cambridge University Press
 testifying understanding

far less clear whether one can gain the relevant skills or abilities required for understand-
ing in the same way, or whether the fact that one can acquire information directly through
testimony means that one can also acquire understanding directly through testimony. We
have also seen that the relevant skills or abilities are often spelled out in terms of a kind of
“grasping” between bits of information. The grasping aspect of understanding seems to be
the most signicant obstacle to acquiring understanding directly through testimony: while
I could, presumably, acquire all of the groundwork for understanding in the form of the
relevant pieces of knowledge from testimony, the grasping component of understanding is
not, seemingly, something that I can acquire through testimony. I turn to this problem
next.

2.5 What is the nature of grasping?

The nal major open question about understanding concerns the nature of its requirement
that to understand one must “grasp” some information or the relationship between some
information. When grasping is referred to, however, it is typically done so metaphorically,
with little attention having been paid to analysing it in any detail. Consider, for example,
the following sampling of some discussions of grasping:

Understanding involves an already-possessed awareness of the explanatory and other connections

involved in the subject-matter in question, an already-mastered grasp that involves or generates the
illumination of a subject we resort to the language of intelligibility and sense-making to convey . . .
the features that constitute the nature of understanding are factivity plus the grasping of internal,
structural relationships between pieces of information. (Kvanvig 2003: 6)

[Understanding] involves grasping relations of parts to other parts and perhaps the relation of
parts to a whole . . . Unlike beliefs, understanding is not passed along from testier to a recipient.
The person’s own mind has to do the “work” of understanding. (Zagzebski 2008: 144–5)

[Y]ou cannot understand why p unless p is true and you have the appropriate grasp of the reasons
why p . . . The grasp of the reasons why p that is essential to understanding involves a number of
abilities: to understand why p, you need to be able to treat q as the reason why p, not merely
believe or know that q is the reason why p. (Hills 2009: 6)

[To mentally] grasp how the different aspects of a system depend upon one another is to be able to
anticipate how changes in one part of the system will lead (or fail to lead) to changes in another
part . . . “Grasping” a structure would therefore seem to bring into play something like a modal
sense or ability – that is, an ability not just to register how things are, but also an ability to antici-
pate how certain elements of the system would behave, were other elements different in one way or
another. (Grimm 2006: 11–12)

While the above conceptions of grasping are predominantly metaphorical, some have
recently proposed more systematic conceptions. For example, Kareem Khalifa (2013)
argues that:

[G]rasping involves reliable explanatory evaluation . . . Explanatory evaluators’ inputs are various
potential explanations of a phenomenon plus a body of relevant evidence, and their outputs are beliefs

[Link] Published online by Cambridge University Press

e p i s t e m e v o l u m e 14–1 113
 kenneth boyd

about which of these potential explanations is an actual explanation of this phenomenon . . .

Explanatory evaluators are reliable just in case the resulting beliefs could not easily have been false. (6)

Hills (Forthcoming) also provides a detailed conception of grasping:

When you grasp a relationship between two propositions, you have that relationship under your
control. You can manipulate it. You have a set of abilities or know-how relevant to it, which you
can exercise if you choose. For instance, if you understand why p, you can give an explanation of
why p and you can do the same in similar cases. If you nd out that q (where q is why p), you can
draw the conclusion that p (or that probably p, if q only makes p probable). And you can do the
same in similar cases. (3)

Hills goes on to list six “cognitive control” abilities involved with “understanding why p
(and q is why p)”, such that one can:

(i)follow some explanation of why p given by someone else.

(ii)explain why p in your own words.
(iii)draw the conclusion that p (or that probably p) from the information that q.
(iv) draw the conclusion that p’ (or that probably p’) from the information that q’
(where p’ and q’ are similar to but not identical to p and q).
(v) given the information that p, give the right explanation, q.
(vi) given the information that p’, give the right explanation, q’. (3)

Although each of these engagements with grasping is different, it seems that we can draw
some general lessons from them: specically, grasping seems to involve the exercise of
some kind of cognitive ability or abilities, where instances of grasping involve the relation-
ships between the thing understood and either the reasons that make that thing true or
propositions that are related to it.
Other authors discuss certain abilities that must be present in order to have understand-
ing without discussing these abilities in terms of “grasping.” Wilkenfeld (2013), for
example, argues that understanding requires “an ability to manipulate some mental cor-
relate of the understood object such that, in the absence of interfering factors, one would
then be able to manipulate the target itself” (1003), while de Regt and Dieks (2005) posit
as a criterion of understanding that one be able to “recognize qualitatively characteristic
consequences of [theory] T without performing exact calculations” (151). It is not clear
whether all of these abilities are necessarily cognitive abilities: one might have a concep-
tion of grasping, or of the abilities required for understanding generally, that one be
able to do something practical (as is perhaps the case in instance of understanding-how),
or that one have an ability that is domain-specic (for example, in the case of scientic
understanding, one might have to be able to do specically scientic things, such as mak-
ing some information visualizable, or to be able to model it effectively, etc.; see, for
instance, de Regt and Dieks 2005 and Lenhard 2009). Although these conceptions of
the skills and abilities required for understanding are different, they will ultimately create
the same problems for the view that understanding can be acquired through testimony,
namely that while understanding requires the exercise of some kind of ability, exercising
that ability prevents one from being able to acquire understanding directly through testi-
mony. In what follows, then, I will address this concern in terms of grasping.

114 e p i s t e m e v o l u m e 14–1
[Link] Published online by Cambridge University Press
 testifying understanding

That understanding requires a kind of grasping has a number of consequences concern-

ing the ability to acquire it through testimony. One consequence of the grasping require-
ment seems to be gaining understanding is something that has to be primarily attributable
to the subject, since grasping is something that the subject has to do for themselves. When
it comes to gaining knowledge, by contrast, it seems that the knowledge that one gains can
be primarily attributed to someone other than the subject. Consider the following case:

Jenny the Tourist: Jenny arrives at the train station in Chicago and, wishing to obtain directions to
the Sears Tower, approaches the rst adult passer-by that she sees. Suppose further that the person
that she asks has rst-hand knowledge of the area and gives her the directions that she requires.
Intuitively, any true belief that Jenny forms on this basis would ordinarily be counted as knowl-
edge. (Pritchard et al. 2010: 40; attributed to Lackey 2007)

There are interesting lessons to be drawn from the Jenny case: Lackey (2007) argues that it
shows that one can have knowledge that is not at all creditable to the knower, while
Pritchard (Pritchard et al. 2010) weakens the claim by arguing that the case shows that
one can have knowledge that is not primarily creditable to the knower. Regardless, by
acquiring knowledge in this way one is more of a passive receptacle than an active partici-
pant: the majority of the cognitive work that was required in coming to know where the
Sears Tower is located was done primarily by the testier, and they are merely bestowing
their knowledge to Jenny.
It might be difcult to see how one could acquire understanding in the way that Jenny
acquires her knowledge. Returning to the example with which we started, while it seems
that you can gain a lot of knowledge about baseball through my testimony, what you can-
not gain through my testimony is a grasp of the relevant connections between the bits of
knowledge that you have gained. Again, while I can help you lay the groundwork, the fact
that understanding is something that has to be primarily attributable to you means that I
cannot be entirely or primarily credited for your understanding. This would imply, then,
that there is at least one major component of understanding that cannot be acquired
through testimony.
However, I will argue that even though one cannot transmit or acquire a “grasping”
via testimony, there are kinds of understanding for which the mechanisms that are
required for grasping are ones that are activated simply by processing information con-
veyed through testimony. Just as knowledge can be successfully acquired through testi-
mony if the hearer possesses concepts and background information that are required to
process the relevant information in the right way, so too can understanding be successfully
acquired through testimony if the hearer possesses the concepts and background informa-
tion that are required to grasp the relevant information in the right way.

3. easy and easys understanding

Some things are difcult to understand – how the large hadron collider works, the behav-
ioral patterns of cuttlesh, the game of cricket, etc. – while other things are easy to under-
stand – how a pencil works, the behavioral patterns of a housecat, the game of tic-tac-toe,
etc. Consider the following sketch of a difference between these kinds of understanding:
difcult understanding requires extensive cognitive effort, potentially over the course of

[Link] Published online by Cambridge University Press

e p i s t e m e v o l u m e 14–1 115
 kenneth boyd

long periods of time, and the implementation of a wide range of cognitive skills and abil-
ities and application of background beliefs and knowledge; easy understanding does not
require nearly as much, if any, cognitive effort, nor does it require the possession of any
unique or well-developed cognitive skills or abilities or any specialized background beliefs
or knowledge. There may be more or fewer characteristics of these kinds of understanding,
and the distinction certainly does not represent a strict dichotomy. Nevertheless, the point
is that some things are simply easier to understand than others.
This distinction between easy and difcult understanding is one that exists regardless of
what we take understanding’s paradigmatic form to be. Consider the following examples:

Understanding-why Russell’s paradox is a reason to reject naïve set theory is difcult;

understanding-why you left the house early this morning to teach your set theory class is easy
(there is construction and you wanted to beat the trafc).

Understanding-how to program a videogame is difcult; understanding-how to play the video-

game Pong is easy (“move the paddle to hit the square: got it”).

Understanding-□ the Schrodinger’s cat thought experiment is difcult; understanding-□ my cat is

easy (she is motivated by cat food and sleep).

In all of the above cases we can assume a layperson’s familiarity with the relevant subject
matters; thus understanding-why Russell’s paradox is a reason to reject naïve set theory is
difcult for the layperson who is not familiar, or only has limited familiarity, with logic or
set theory, whereas understanding-why you left the house early this morning is easy for
the layperson who has only the most basic of relevant background information, etc.
Discussions of understanding often focus just on those kinds of understanding that are
difcult; after all, these seem to be the most interesting and – a point I will return to below
– valuable cases of understanding.13 Furthermore, difcult understanding does, in fact,
require signicant cognitive effort in the form of conscientious grasping. That these factors
should stand in the way seems to be just what is going on in the above cases of difcult
understanding: to understand-why Russell’s paradox is a reason to reject naïve set theory
seems to require more, say, than just my giving you a proof of Russell’s paradox, since you
also have to grasp how the parts of the proof relate to one another and to the debate as a
whole; understanding-how to program a videogame requires signicant study, practice,
and the development of cognitive skills and abilities; and understanding-□ the
Schrodinger’s cat thought experiment almost certainly requires more than just reading
the relevant chapter from a quantum mechanics textbook.
What about cases of easy understanding? Acquiring any of the instances of easy under-
standing in the above examples does not seem to require signicant cognitive effort or the
development of any specic kinds of skills or abilities, cognitive or otherwise: these are
instances of understanding that can be acquired by anyone with only the most basic of

13 Consider, for example, one of Kvanvig’s examples of understanding: “a historical understanding of the
Comanche dominance of the southern plains of North America from the late seventeenth until the late
nineteenth century” (2003: 197), or Pritchard’s examples of “combustion or quantum mechanics or
Republican ideology” (in Pritchard et al. 2010: 104), or any of the examples of objects or phenomena
listed in the previous section.

116 e p i s t e m e v o l u m e 14–1
[Link] Published online by Cambridge University Press
 testifying understanding

background familiarity with the relevant subject matter. Of course, there needs to be some
familiarity with the relevant background information: if you have absolutely no familiar-
ity with even the most rudimentary of videogames, for example, you might not be able to
understand-how to play Pong without gaining some additional knowledge rst. But this
should not be surprising: any successful testimonial act will require at least some back-
ground familiarity with the relevant subject matter. Consider again the traditional picture
of how knowledge is acquired through testimony. First, I know that p and I tell you that p.
You can then come to know that p on the basis of my telling you, but there are certain
conditions that need to be in place in order for you to gain that knowledge: I need to
have knowledge (say14), you need to trust that I know what I’m talking about (at least
implicitly, and either for positive reasons or lack of negative ones), and you have to
have some background beliefs or concepts that allow you to properly interpret the content
of what I am saying (otherwise you could at best parrot what I was saying, but you would
not know it). But acquiring knowledge from me does not necessarily involve any signi-
cant cognitive effort or the implementation of any specialized skills or abilities: you do
not, say, need to explicitly go through the process of reasoning that concludes, from
the basis of my trustworthiness, combined with facts surrounding p, etc., that you should
accept p, and only thereby come to know it, nor do you need to be able to manipulate the
information that p in any particular way, or anticipate consequences regarding closely
related propositions, etc.
The acquisition of knowledge, then, can be primarily a passive affair. However, there
still has to be something going on in the mind of the recipient of testimonial information in
order to gain knowledge, this is just not something that is necessarily particularly cogni-
tively effortful, or the result of any specialized abilities, or that involves the application of
any particular background information or knowledge, etc. Acquiring beliefs or knowledge
does require that we process the information: we have to make sense of the content of
one’s testimony, and we have to conceive of the testier or the content of her testimony
in the right way, but that is all. Now, what this “right way” consists in is a matter of sign-
icant debate. As we have seen briey above, two main views about testimony as a source
of knowledge state that we either need to possess positive reasons to believe an individual,
either in a given circumstance or as a source of testimony more generally, or that one lacks
reason to disbelieve or doubt the source of the information or the veracity of the informa-
tion itself.15 Different views require more work on behalf of the hearer than others.
Consider, for example, the following views. Robert Audi (1997), argues that:

[B]eliefs about the credibility of the attester and beliefs pertinent to the attested proposition play a
mainly ltering role: they prevent our believing testimony that does not “pass,” for instance
because it seems insincere; but if no such difculty strikes us, we “just believe” (non-inferentially)
what is attested. These ltering beliefs are like a trap door that shuts only if triggered; its normal
position is open, but it stays in readiness to block what should not enter. The open position is a
kind of trust. (406)

14 I am assuming here, for argument’s sake, something like the traditional “transmitting” model of tes-
timony; nothing substantial hangs on this assumption.
15 These views correspond to the local and global versions of reductionism, and non-reductionism,
respective. See, for discussion and critique, Lackey (2008: Ch. 5).

[Link] Published online by Cambridge University Press

e p i s t e m e v o l u m e 14–1 117
 kenneth boyd

Elizabeth Fricker’s (1994) view requires perhaps slightly more work on behalf of the
hearer:

[T]he hearer should be discriminating in her attitude to the speaker, in that she should be continu-
ally evaluating him for trustworthiness throughout their exchange, in the light of the evidence, or
cues, available to her. This will be partly a matter of her being disposed to deploy background
knowledge which is relevant, partly a matter of her monitoring the speaker for any tell‐tale
signs revealing likely untrustworthiness. (150)

Goldberg and Henderson (2006) argue for a similar requirement, that one “monitor” the
quality of incoming information: they compare the kind of monitoring they have in mind
to a “buzzer system designed to exploit testimonial regularities” (610). Finally, Lackey
(2008) argues that, generally, “the hearer must also do her part in a testimonial exchange
by having at least some epistemically relevant positive reasons for accepting the report in
question” (175).
Different conceptions of testimony, then, require the hearer to do different amounts of
cognitive work in order to acquire knowledge. Regardless of the amount of work one
needs to do, however, one is still able to acquire knowledge directly through testimony:
testimony does not merely lay the groundwork for the acquisition of knowledge, one
acquires knowledge via testimony. The case appears to be very different in discussions
of the acquisition of understanding: as we have seen above in the various passages regard-
ing grasping, understanding requires active cognitive work on behalf of the understander,
and the exercise of specic cognitive abilities. The standard view concerning the acquisi-
tion of understanding, then, is that understanding requires much more than processing
information (as we have seen, it is the extra cognitive work that is taken to stand in the
way of understanding being acquired through testimony). The purpose of introducing
the idea of easy understanding is to challenge the standard view, by showing that there
are instances in which we can set up the conditions for gaining easy understanding in
the same way as we can gain knowledge: if I have understanding, you possess the relevant
background beliefs and abilities that allow you to properly interpret what I’m saying, and
you conceive of me or the content of my testimony in the right way, then when the content
of what I’m testifying is easy to understand, you plausibly do not need to do any more
cognitive work to gain understanding than is required to gain knowledge. Since processing
does not stand in the way of acquiring beliefs or knowledge directly through testimony,
nor should it stand in the way of acquiring understanding directly through testimony.
The idea, then, is the following: in cases of easy understanding, all of the mechanisms
that are needed to grasp the relevant information, or connections between the relevant bits
of information, are already activated when processing the relevant information. By posses-
sing the relevant background information and concepts, then by successfully processing
my testimony that, e.g., “to play Pong, move the paddle to hit the square,” you have there-
by done enough to grasp the relevant information or connections between bits of informa-
tion, namely those connections between the information that I am conveying to you
through testimony and the background information that you already possess. Even though
an instance of understanding involves both an informational component and a grasping
component, then, one can “slot in” new information into an already-possessed web of
grasped relationships between different pieces of information. Thus while one still needs
to full the grasping requirement to gain new understanding, that grasping is not

118 e p i s t e m e v o l u m e 14–1
[Link] Published online by Cambridge University Press
 testifying understanding

something that can be acquired through testimony does not prevent one from being able
to acquire new understanding through testimony, since one can gain new understanding
by slotting in new information to already-grasped relationships between information.16
Consider Kvanvig’s (2003) discussion of grasping, whereby grasping involves the
“internal, structural relationships between pieces” of the relevant information. In the
case of easy understanding, the relationships between the relevant pieces of information
are so easy to grasp that by successfully processing the content of what I am saying,
and given your background knowledge and familiarity with the subject matter, you
have done enough to grasp the relevant relationships. Again, consider the Pong case: if
your background knowledge and familiarity with the relevant subject matter is enough
such that you can make sense of what I mean when I say “move the paddle to hit the
square” then by making sense of my utterance you have done enough to understand
the relevant relationships between the bits of information. The lesson to be drawn from
cases of easy understanding is that while easy understanding still requires a kind of grasp-
ing, this grasping does not stand in the way of easy understanding being acquired through
testimony. Some things are just easier to grasp than others, and some things are so easy to
grasp that by successfully processing the information that I have presented to you, you
have already thereby grasped the relationships needed to gain the relevant understanding.
I noted above that even in cases of easy understanding, someone who had no familiar-
ity with the relevant subject matter would not necessarily be able to gain the relevant
understanding without more signicant cognitive effort or without developing some
new kind of ability. How easy it is to understand something, then, is partly a function
of the thing being understood, but also partly a function of the subject. Consider, then,
another, more general kind of understanding: easys understanding. This kind of under-
standing is relativized to the subject: given S’s background familiarity, knowledge, and
cognitive skills and abilities, some things are easier to understand for S than others.17
One need only introspect on one’s own strengths and weaknesses to nd ready examples
of things that are easy for one to understand and things that are difcult for one to under-
stand: for example, understanding how to prove a theorem in predicate logic is easy for
me, but it is difcult for my students; understanding the aesthetic advantages of the cur-
rent generation of iPhone over the previous versions is difcult for me, but easy for my
students; etc.

16 The same can be said for those conceptions of understanding that require that one possess certain prac-
tical or domain-specic abilities: when acquiring new scientic understanding, for example, one does
not have to develop new skills for every bit of new understanding one gains. Rather, one can apply
previously developed skills to new information. Or consider an instance of gaining new
understanding-how: if one understands-how to throw out a batter sliding into second base, then
one can come to understand-how to throw out a batter sliding into third base by gaining some new
information, but without having to develop new skills. The picture of gaining new understanding
by slotting in new information into a network of cognitive skills and abilities is the same when it
comes to practical or domain-specic skills: by combining new information with old skills, one can
come to acquire new understanding.
17 All instances of easy understanding are thereby instances of easyS understanding, where the kind of
background knowledge and cognitive skills needed for easy understanding are those that are taken
for granted in a particular situation or context by the majority of people and in the majority of every-
day contexts.

[Link] Published online by Cambridge University Press

e p i s t e m e v o l u m e 14–1 119
 kenneth boyd

Given a sufcient amount of background familiarity, knowledge, and skills, there are
some things that are very difcult for the vast majority of people to understand, but
which are nevertheless easy for some particular individuals to understand. For instance,
if someone is an expert in a given eld, there might be things that are related to this
eld that are easy for the expert to understand but that might be very difcult for a novice
to understand. As was the case for easy understanding, there are instances of easys under-
standing regardless of what we take understanding’s paradigmatic form to be. Consider
the following examples:

Understanding-why the butler did it is difcult for Watson, but easy for Sherlock Holmes.

Understanding-how to perform the latest hypermodern defense is difcult for the chess novice, but
easy for the chess master.

Understanding-□ the Riemann hypothesis is difcult for the mathematics undergraduate, but easy
for the mathematics professor.

In all of the above cases the respective instances of understanding would be difcult for a
novice – it would have to be obtained through conscious cognitive effort, and would, no
doubt, require the development of new skills and abilities to be able to grasp the relevant
concepts and relationships. But since Sherlock Holmes is an expert in reasoning, it is easy
for him to understand-why the butler did it; since the chess master has so much experience
playing chess and studying chess strategy, it is easy for her to understand-how to perform
the latest hypermodern defense; since the mathematics professor has such a strong mind
for mathematics, it is easy for them to understand-□ the Riemann hypothesis.
Easy understanding involves understanding for which most people possess the relevant
background familiarity, knowledge, and cognitive skills, but for which they did not have
to conscientiously develop, and I argued above that once these background conditions are
in place, one can potentially gain easy understanding through testimony since the grasping
that is required for understanding can occur automatically at the level of processing the
testied information. In cases of easys understanding, the story is the same: although
the kind of familiarity, knowledge, and skills that an expert has regarding a given subject
matter are ones that require a signicant amount of conscientious cognitive effort to
acquire, once they have been acquired that which would be difcult for a novice to under-
stand becomes easy for the expert to understand.
Consider, for example, how one expert might communicate understanding to either a
novice or another expert. In the former case, it may very well be that all the expert can do
is lay the relevant groundwork: they can, to the best of their ability, communicate the
beliefs and knowledge to the novice that will ultimately form the basis of understanding,
but there is still signicant work that the novice needs to do in order to acquire that under-
standing for themselves. When an expert testies something that they understand to
another expert in their eld, however, the relevant groundwork has already been laid,
and thus no signicant cognitive effort needs to be expended on behalf of the hearer.
Again, consider the underlying mechanisms that are necessary for the transmission of
knowledge: one cannot acquire knowledge that p through testimony if one does not pos-
sess at least some background beliefs or concepts that can allow one to make sense of the
truth of p. Once these concepts and cognitive abilities are in place, the acquisition of

120 e p i s t e m e v o l u m e 14–1
[Link] Published online by Cambridge University Press
 testifying understanding

knowledge can be easy. And although there is clearly something that is going on internally
when knowledge is acquired – I cannot, for instance, form your belief for you – this
requirement does not stand in the way of knowledge being directly acquirable through tes-
timony. Of course, you do not have to actively form such a belief, in the sense of expend-
ing any signicant cognitive effort, and indeed you might not have any control over
whether you gain the relevant belief or knowledge at all. But the same seems to be the
case in those cases of easy understanding and easys understanding: although whatever
grasping takes place has to be something that the hearer does, this is not enough to
stand in the way of directly acquiring the relevant understanding.
Of course, this is not to say that all instances of understanding can be acquired through
testimony, or that there is any instance of understanding that all people can acquire
through testimony: difcult understanding, and that which is difcult for one to under-
stand, presumably, cannot always be acquired in this way. One general category of under-
standing that is difcult to acquire is that which pertains to an entire subject matter: an
example that we have seen is that of understanding quantum mechanics, something
that is difcult not only because the subject matter is complex, but because of the quantity
of information that is needed to be grasped. However, we have also seen that not all
instances of understanding have to be so large in scope: while instances of understanding
might not be so discrete as to pertain to individual propositions, there are still new
instances of understanding that one can gain about specic things. This is perhaps most
straightforward in the case of easys understanding: an expert in a given eld might have
a lot of knowledge, understanding, and, with the relevant cognitive skills and abilities
in place, then gain new understanding concerning some development in their eld. We
would no doubt say that in these situations the expert has gained some new
understanding.18
There is also a very large gray area between “novice” and “expert”: someone who is
somewhat familiar with a given subject matter will have to work harder to gain under-
standing than someone who is very familiar with it, and someone who is not at all familiar
will have to work harder still. Consider again some of the examples discussed earlier: if
someone knows and understands a lot about baseball aside from intentional walks,
then understanding-why the pitcher intentionally walked the batter could be as simple
as merely acquiring new information about baseball strategy, while someone who is
less familiar with baseball might have to do more to put together the pieces. Or consider
an experienced boxer, who has fought right-handed opponents for many years, who then
learns that “beating a southpaw is just like beating a right-hander, except you have to
counter their left instead of their right”: given the extent of the skills that the boxer has
developed, she might not need to do anything more beyond processing this new informa-
tion in order to understand-how to beat a southpaw, whereas a less-experienced boxer,
when presented with the same information, might have to work a lot harder, either by
doing more cognitive work with regards to the new information and what they already
know and understand, or by sparring with a southpaw. Finally, consider a political pundit
who has yet to meet the latest Republican presidential candidate, but is told “he’s just like
the last guy, only even more scally conservative.” Given the extent of the pundit’s

18 We would probably also say that the expert also now has a better understanding of their specic eld.
But to say that one has gained some new understanding and that one has a better understanding of
something they had some understanding of before are not incompatible statements.

[Link] Published online by Cambridge University Press

e p i s t e m e v o l u m e 14–1 121
 kenneth boyd

experience and expertise, processing this information seems sufcient to give him new
understanding, while a less-experienced political analyst would have to do more, either
in the form of gaining more information or conscientiously applying cognitive or other
skills, to gain the same understanding.
I have argued that in all of the above cases, experts can acquire new understanding sim-
ply by acquiring new information, and since new information can be acquired directly
through testimony, so too can new understanding: in at least some cases of easys under-
standing, the only thing preventing someone from understanding-why something is the
case, how something is done, or some object or phenomenon, is the relevant informational
component. Again, it still remains the case that when receiving this new information, the
subject must do something with it to gain understanding. But the same is the case when
gaining new knowledge as the result of gaining new information: the mere fact that some-
thing has to go on in the mind of the subject when gaining knowledge does not mean that
one does not acquire one’s knowledge directly through testimony. What would stand in
the way of one acquiring an epistemic relationship directly through testimony would be
if there had to be some signicant or active cognitive work on the part of the subject.
But when something is easy or easys for someone to understand, gaining that new under-
standing need not be any more cognitively taxing than when one gains new knowledge.
The purpose of introducing the notions of easy and easys understanding was to show
that the grasping component of understanding does not itself have to be acquired through
testimony in order to gain understanding; rather, the informational component of a given
instance of understanding can be acquired through testimony and grasped with structures
and abilities that are already in place. The grasping component on its own, then, does not
prevent understanding from being acquirable through testimony; rather, what might pre-
vent an instance of understanding from being acquirable through testimony is that the
hearer does not possess the relevant structures or cognitive abilities needed to do the
requisite grasping.
One might, however, think that there is an aspect of understanding that I am ignoring,
one that may very well prevent it from being acquirable through testimony. I turn to this
next.

4. understanding, achievement and value

One of the motivations behind discussing understanding in the rst place is that it seems to
have value. While knowledge has in recent history been taken to be the primary bearer of
epistemic value, it has recently been called into question whether knowledge has any value
over and above its traditionally held component parts (such as truth, justication, belief,
and some appropriate de-Gettierizing condition) either collectively or individually.19 If
knowledge is not uniquely valuable, the reasoning goes, then perhaps we should be look-
ing for an epistemic relationship that is, and understanding seems to be a prime candidate.
Why might understanding be uniquely valuable? One reason that has been given is that
understanding requires a certain kind of achievement, one that is not required by knowl-
edge. For example, Pritchard (Pritchard et al. 2010) distinguishes two different kinds of

19 For an overview of some of these issues, see Riggs (2008), and Pritchard et al. (2010).

122 e p i s t e m e v o l u m e 14–1
[Link] Published online by Cambridge University Press
 testifying understanding

achievements, what he calls weak achievements and strong achievements: weak achieve-
ments are “successes that are because of ability,” whereas strong achievements are “suc-
cesses that are because of ability where the success in question either involves the
overcoming of a signicant obstacle or the exercise of a signicant level of ability” (70).
Pritchard argues that while on some conceptions of knowledge, knowledge requires a cog-
nitive achievement on the part of the knower, it is at most a weak cognitive achievement;
understanding, on the other hand, requires a strong cognitive achievement.
Let us grant, for the time being, that understanding requires a strong cognitive achieve-
ment in Pritchard’s sense. One might then have the following objection to my conceptions
of easy understanding and easys understanding: since, in none of the cases I describe above
does the hearer exhibit a strong cognitive achievement, they do not count as cases of
understanding at all. Consider again the case of Jenny the Tourist: if Jenny were to
have achieved anything at all, it would be at best a weak kind of achievement. This
kind of achievement might be good enough for her to know where the Sears Tower is,
but since she did not overcome any signicant obstacles or exercise her abilities in any
signicant way, her achievement would not be good enough to qualify her as having
understanding. Is the story not the same when it comes to easy understanding?
Consider again how I might tell you how to play the videogame Pong: “move the paddle
and hit the square.” After hearing this we can now sit down together and play Pong, but
what have you really gained from me as a result of my testimony? You have not, it seems,
had to overcome any signicant obstacles, nor did you exercise your cognitive abilities in
any signicant way. If you have not achieved anything in the strong sense as a result of our
interaction, then perhaps you do not have any understanding. Strong cognitive achieve-
ments might then be incompatible with testimony generally: acquiring information
through testimony, we might think, is something that is by its very nature easy, and
does not require any signicant cognitive achievement. That is not to say that one cannot
learn new and difcult things through testimony, but again, the testimony itself serves
merely to lay the relevant groundwork: signicant cognitive achievements are things
that the subjects have to achieve for themselves.
While this line of argument is one that would provide a principled reason why under-
standing cannot be acquired through testimony, I do not think it is successful. To see why,
consider those strong achievements that are nevertheless easy for someone: Pritchard’s
own example is one of Tiger Woods sinking a long putt (Pritchard et al. 2010: 68).
Sinking a long putt is very difcult for me, but it is easy for Tiger Woods; however,
when Tiger Woods sinks a long putt, the ease with which he does so, argues Pritchard,
does not diminish his achievement. We can still credit Tiger Woods with an achievement
in the strong sense because he is nevertheless exercising a signicant ability, and while he is
not overcoming a signicant obstacle in the very moment in which he makes the putt, he
doubtless had to overcome such obstacles while he was developing his skills in the past. In
general, something that can be easy for someone can still qualify as an achievement in
Pritchard’s strong sense.
The same, however, seems to be the case when it comes to cognitive achievements:
being an expert in something – possessing the relevant background familiarity, knowledge,
and skills – makes it easy for one to understand things that pertain to that person’s area of
expertise, just as sinking a series of long putts is easy for Tiger Woods. And just as we
should not consider Tiger Woods’ putts to be achievements in the weak sense simply
because they are easy for him, nor should we consider the expert’s cognitive achievements

[Link] Published online by Cambridge University Press

e p i s t e m e v o l u m e 14–1 123
 kenneth boyd

any weaker simply because they are easy for her: in gaining understanding the expert still
exercises a signicant ability, and in exercising that ability displays the overcoming of a
signicant obstacle (even if, as is the case with Tiger Woods, this obstacle was one that
was overcome not in the present instance, but in the development of one’s abilities in
the past). Something that is easyS to understand, then, does not disqualify it from being
a cognitive achievement in the strong sense, and thus does not disqualify it from being
an instance of understanding.20
We also saw earlier that when it comes to acquiring knowledge through testimony, one
can gain knowledge in a way that is not primarily attributable to the knower: in Jenny the
Tourist, Jenny gains the relevant knowledge, but whatever kind of cognitive achievement
occurs in such an exchange is not one that is primarily attributable to Jenny. In the case of
gaining easyS understanding through testimony, however, the relevant cognitive achieve-
ment is still attributable to the understander: it is, after all, their hard work that has
put them in the position to be able to acquire that understanding in the rst place.21
Ease of acquisition, then, is incompatible with neither signicant achievement nor being
attributable to the acquirer.
Good, then, for easyS understanding, but what about easy understanding? One might
wonder whether easy understanding should qualify as understanding at all, given that it
does not seem to require a signicant cognitive achievement. Or, consider what
Wilkenfeld et al. (Forthcoming) have argued, namely that “understanding-why has higher
demands for knowledge-why when it comes to explanatory depth” (4), where one is able
to provide a deep explanation if one is able to see connections between pieces of relevant
information, extrapolate from relevant causes and effects, amongst other requirements.22

20 One might wonder whether we should, in fact, consider Tiger Woods’ putts to be achievements in the
strong sense; after all, they are still easy for him, and while we might consider it to be quite the feat to
sink a long putt, Tiger himself might not conceive of it as anything special. I do not have any stake in
whether Pritchard’s conception of strong cognitive achievements is correct; I merely assume it here for
the sake of argument. However, if we were to think that Tiger Woods’ putts did not count as achieve-
ments in the strong sense, Pritchard’s view might run into trouble with the cognitive analog: we would
have to say that an expert’s acquisition of new information that is easy for them would not qualify as a
strong cognitive achievement, and would thus not qualify as an instance of understanding. As a result,
we would have to say that experts are unable to acquire hardly any new understanding in the relevant
eld. Furthermore, the more expertise in the form of skills, abilities, and knowledge that an expert
gained, the less new understanding they would be able to acquire, as their additional expertise
would preclude the possibility of being able to cognitively achieve in the strong sense. This is, I
think, somewhat counterintuitive: experts, it seems, acquire new understanding in their eld all the
time, and gaining new expertise should result in one being able to acquire understanding more easily,
not stand in the way of acquiring new understanding at all. Again, while it might seem tempting to say
that Tiger Woods’ long putt does not qualify as a strong achievement because he is so good at golf, it is
much less tempting to say, for example, that a physics professor’s acquisition of new information does
not qualify as understanding because she is so good at physics. Thanks to an anonymous referee for
suggesting I address this objection.
21 One might argue that we can similarly attribute an achievement to Jenny: she did, after all, successfully
seek out someone to ask, and possesses the relevant background beliefs and knowledge to be able to
make sense of the passerby’s testimony. We might think, however, that even if we should attribute
some kind of achievement to Jenny, it is not one that involves the overcoming of a signicant obstacle,
either in the instance or before.
22 Wilkenfeld et al. do not provide an explicit theory of “explanatory depth”; the version I present here is
based on a collection of remarks from the authors’ (Forthcoming). For a more detailed examination of
explanatory depth, see Ylikoski and Kuorikoski (2010).

124 e p i s t e m e v o l u m e 14–1
[Link] Published online by Cambridge University Press
 testifying understanding

One might think, then, that in purported cases of easy understanding since one could only
provide a shallow explanation for why relevant phenomenon occurred, then one does not
actually have understanding in these cases at all.23
Above I assumed Pritchard’s view for the sake of argument to see if requiring a sign-
icant cognitive achievement would provide a principled reason why understanding
could not, in general, be acquired through testimony. If, according to this picture, easy
understanding does not qualify as a legitimate kind of understanding, then my argument
goes through unharmed, since my argument applies to cases of easyS understanding, and
cases of easy understanding will simply not count. The same can be said for a view accord-
ing to which understanding-why requires signicant explanatory depth: if what I have pre-
sented as cases of easy understanding fail to be legitimate cases of understanding, then
they do not affect my arguments concerning legitimate cases of understanding.
Furthermore, cases of easyS understanding will be able to meet the explanatory depth
requirement: because of an expert’s background knowledge and abilities, they will be
able to provide deep explanations for why some new information gained through testi-
mony is the case. Again, as was the case with signicant cognitive achievements, one
does not have to acquire the knowledge and abilities that can provide explanatory
depth all at once.
I stated earlier, however, that discussions of understanding tend to focus on instances
of understanding that are difcult, and that it was partially as a result of this focus that
made it seem as though there was something about understanding generally that prevents
it from being acquired through testimony. A similar problem occurs in discussions of the
nature of understanding: the cases I have listed as instances of easy understanding are, I
think, natural and intuitive, and there is, to my mind, no antecedent reason why certain
instances of understanding cannot be simple, or why we cannot understand things that,
by their very nature, only admit of shallow explanations. To show otherwise is a burden
of proof that those writing on understanding need to assume.
We saw that one of the reasons why we might conceive of understanding as more cog-
nitively demanding, or more difcult to acquire than knowledge, is that we can more eas-
ily attribute to it a special kind of epistemic value. If understanding can, in fact, be
acquired through testimony, does this affect its epistemic value? I don’t think so. Just as
the fact that some things are easy for some people does not, in itself, decrease that thing’s
value, neither does the fact that some understanding is easy for some people to acquire
decrease the value of that understanding. The fact that knowledge is sometimes easy to
acquire through testimony has been taken to be an indication that knowledge is not par-
ticularly epistemically valuable since, as cases like Jenny the Tourist show, one can acquire
knowledge with very little effort on one’s part. However, it seems that drawing general
conclusions about the value of an epistemic relationship from individual instances in
which one can acquire it easily is a mistake: it is likely that for any given subject matter
and any epistemic relationship one can have with it, there will be someone, somewhere,
for whom acquiring that relationship with that subject matter will be easy.
None of this is to deny that some things are more difcult to understand than others.
Nor does it deny that there are some instances of understanding that cannot be acquired
through testimony by some subjects: if something is difcult for a subject to understand,

23 Thanks to an anonymous referee for suggesting I address this objection.

[Link] Published online by Cambridge University Press

e p i s t e m e v o l u m e 14–1 125
 kenneth boyd

then it may very well be the case that all that that subject can acquire through testimony is
the groundwork for understanding. However, I have argued that none of the components
of understanding – its relationship to knowledge, factivity, susceptibility to luck, paradig-
matic form, or grasping – stand in the way of being able to acquire understanding through
testimony. In those instances in which one is unable to gain some understanding through
testimony, then, what stands in the way is that it is difcult for the subject, not that it is an
instance of understanding.24

references
Audi, R. 1997. ‘The Place of Testimony in the Fabric of Knowledge and Justication.’ American
Philosophical Quarterly, 34(4): 405–22.
Brogaard, B. 2009. ‘Why Mary Did Yesterday: Reections on Knowledge-Wh.’ Philosophy and
Phenomenological Research, 78(2): 439–67.
de Regt, H. and Dieks, D. 2005. ‘A Contextual Approach to Scientic Understanding.’ Synthese,
144(1): 137–70.
Elgin, C. 2007. ‘Understanding and the Facts.’ Philosophical Studies, 132: 33–42.
——— 2009. ‘Is Understanding Factive?’ In D. Pritchard, A. Millar, and A. Haddock (eds), Epistemic
Value, pp. 322–30. Oxford: Oxford University Press.
Fricker, E. 1994. ‘Against Gullibility.’ In B. K. Matilal and A. Chakrabarti (eds), Knowing from
Words, pp. 125–61. Dordrecht: Kluwer Academic Publishers.
Gettier, E. 1963. ‘Is Justied True Belief Knowledge?’ Analysis, 23: 121–23.
Goldberg, S. and Henderson, D. 2006. ‘Monitoring and Anti‐Reductionism in the Epistemology of
Testimony.’ Philosophy and Phenomenological Research, 72: 600–17.
Goldman, A. 1976. ‘Discrimination and Perceptual Knowledge.’ Journal of Philosophy, 73(20):
771–91.
Grimm, S. 2006. ‘Is Understanding a Species of Knowledge? ’ British Journal of the Philosophy of
Science, 57: 515–35.
——— Forthcoming. ‘How Understanding People Differs from Understanding the Natural World.’
Philosophical Issues.
Hawley, K. 2010. ‘Testimony and Knowing How.’ Studies in the History and Philosophy of Science
Part A, 41(4): 397–404.
Hills, A. 2009. ‘Moral Testimony and Moral Epistemology.’ Ethics, 120(1): 94–127.
——— Forthcoming. ‘Understanding Why.’ Noûs, 49(2).
Hopkins, R. 2007. ‘What is Wrong With Moral Testimony?’ Philosophy and Phenomenological
Research, 74(3): 611–34.
Kelp, C. Forthcoming. ‘Understanding Phenomena.’ Synthese.
Khalifa, K. 2013. ‘Understanding, Grasping and Luck.’ Episteme, 10(1): 1–17.
Kitcher, P. 1989. ‘Scientic Understanding and the Causal Structure of the World.’ In P. Kitcher and
W. Salmon (eds), Scientic Explanation, pp. 410–505. Minneapolis, MN: University of Minnesota
Press.
Kvanvig, J. 2003. The Value of Knowledge and the Pursuit of Understanding. Cambridge: Cambridge
University Press.
Lackey, J. 2006. ‘Knowing from Testimony.’ Philosophy Compass, 1(5): 432–48.
——— 2007. ‘Why We Don’t Deserve Credit for Everything We Know.’ Synthese, 158(3): 345–61.
——— 2008. Learning from Words: Testimony as a Source of Knowledge. Oxford: Oxford University
Press.

24 Thanks to Diana Heney, Andreas Schmidt, and Hasko von Kriegstein for their detailed and thoughtful
comments on an earlier version of this paper. Thanks also to the audience of my session at the 2015
Joint Session of the Aristotelean Society and Mind Association, as well as an anonymous referee, for
their helpful feedback.

126 e p i s t e m e v o l u m e 14–1
[Link] Published online by Cambridge University Press
 testifying understanding

Lenhard, J. 2009. ‘The Great Deluge: Simulation Modelling and Scientic Understanding’. In H. de
Regt, S. Leonelli, and K. Eigner (eds), Scientic Understanding: Philosophical Perspectives, pp.
169–88. Pittsburgh, PA: University of Pittsburgh Press.
Lipton, P. 2004. Inference to the Best Explanation. New York, NY: Routledge.
Morris, K. 2012. ‘A Defense of Lucky Understanding.’ British Journal for the Philosophy of Science,
63: 357–71.
Pritchard, D., Millar, A., and Haddock, A. 2010. The Nature and Value of Knowledge. Oxford:
Oxford University Press.
Riggs, W. 2008. ‘The Value Turn in Epistemology.’ In V. Hendricks (ed.), New Waves in
Epistemology, pp. 300–23. New York, NY: Palgrave Macmillan.
Stanley, J. and Williamson, T. 2001. ‘Knowing How.’ Journal of Philosophy, 98(8): 411–44.
Wilkenfeld, D. 2013. ‘Understanding as Representation Manipulability.’ Synthese, 190: 997–1016.
———, Plunkett, D. and Lombrozo, T. Forthcoming. ‘Depth and Deference: When and Why We
Attribute Understanding.’ Philosophical Studies.
Ylikoski, P. and Kuorikoski, J. 2010. ‘Dissecting Explanatory Power.’ Philosophical Studies, 148(2):
201–19.
Zagzebski, L. 2008. On Epistemology. Belmont, CA: Wadsworth.

KENNETH BOYD is Visiting Assistant Professor in Philosophy at Dalhousie University. His

primary research interests are in epistemology, specically on questions concerning the
nature of epistemic evaluations and obligations.

[Link] Published online by Cambridge University Press

e p i s t e m e v o l u m e 14–1 127