Final spring pdf qxd

---

16 Bulletin of the American Academy   Spring 2004

ates and their signi½cant others, and I chose him

as my signi½cant other. In the last ten years he

took pleasure in signing his letters “Father of

the Economist.” I tried to reciprocate using

“Son of the Sociologist”–but it didn’t start out

that way. 

When I was a child, my father was surely there

and he taught me many important things, not

least about the stock market, poker, and magic.

Only the magic didn’t stick–I didn’t have the

discipline. As a teenager, I had very little inter-

est in the intellectual or academic life. I was in-

terested in cars. I started building hot rods, if

you can believe it, and racing them. It seemed I

did almost everything that took me as far away

as possible from what he did. As we began to

discuss college, I said, “ I think the General Mo-

tors Technical Institute would be good.” That

didn’t happen, but neither did I apply to Har-

vard–I wouldn’t have gotten in if I had. I went

to Columbia Engineering School and majored

in engineering mathematics, but my plan was

still to be an auto engineer, which was a small

move away from becoming a mechanic. I

should mention that during my sophomore

year at Columbia I took a required course in

English, and I wrote what was to become my

½rst published paper, entitled “The ‘Motion-

less Motion’ of Swift’s Flying Island.” When I

turned it in to my English instructor, for whom

to that point I had done D work, he said, “Well,

this has some possibilities for us; maybe we can

get it published.” Fortunately, I had a father

who knew something about such matters. So 

I asked for his advice and eventually told the

instructor, “No, I think it would be just ½ne if

I wrote the paper alone.” So, I have a D in En-

glish on my transcript, but I also have a paper

published in the Journal of the History of Ideas.

After a few summers working for the Ford

Motor Company, I decided perhaps academia

might not be such a bad choice and so, some

thirty-½ve years ago, I went off to Caltech to

do a Ph.D. in applied math. After ½nishing all

my courses and passing my quali½ers at Cal-

tech, I was all set to do my thesis when I began

to wonder whether water waves in a tank or

plasma physics was right for me. Perhaps it

was the fact that I used to get up every morn-

ing and go down to a local brokerage house for

the 9:30 a.m. opening of the New York Stock

Exchange–6:30 a.m. my time. I would trade

all kinds of securities–stocks, convertible

bonds, warrants, and options–and then go off

to class. One day, I came upon a very bad book

on mathematical economics; it was so bad that

I said to myself, “I can do better.” If I had seen

a book by Bob Solow or Ken Arrow, I might

still be doing fluid mechanics, but instead I

Robert C. Merton is John and Natty McArthur Uni-

versity Professor at Harvard Business School. He has

been a Fellow of the American Academy since 1986.

Robert M. Solow, a Fellow of the American Acad-

emy since 1956, is Institute Professor, Emeritus, at

the Massachusetts Institute of Technology.

Robert C. Merton

R

obert K. Merton was my father for nearly six

decades–and he was my best friend for the last

four. In the early years we talked on the phone

at least once nearly every day. Later, when email

came, he mastered it before I did–it relaxed the

constraint in his being a lark and me an owl, so

our interchanges multiplied. We also edited

each other’s papers, although both relative need

and skill ensured that I could never produce

anything like the amount of red ink on his pa-

pers that he put on mine. I know that, over the

years, many of you came to know what I mean

about that red ink.

We would sit by the ½re in East Hampton in his

last ten or ½fteen years and talk for hours. When

I got that early morning phone call in October

of ’97 saying, “We have something interesting

to tell you,” it was 

rkm

(I never called my fa-

ther 

rkm

but the initials did come in handy in

talking about him to others) who got the next

phone call from me. At the Nobel banquet in

Stockholm there is a center table for the laure-

In Memory of Robert K. Merton

Robert C. Merton and Robert M. Solow

This presentation was given at the 1876th Stated Meeting, held at the House of the Academy on

November 12, 2003.

© Sandra Still

---

Bulletin of the American Academy   Spring 2004    17

decided to switch to economics–a move that

everyone thought was crazy, except my father

who was smiling. You can see where this is

heading: I was now in academe and, further-

more, into one of the social sciences.

My research is very mathematical, very quan-

titative, and seemingly very far away from what

my father did. Nevertheless I found myself us-

ing many of his ideas. I will list a few examples.

There is the concept of post-factum interpreta-

tion after the observation to provide an expla-

nation, which he paid a lot of attention to and

was pretty derisive about. Of course, in eco-

nomics, post-factum interpretations are made

daily: read The Wall Street Journal today and you

can learn exactly why the market did what it

did yesterday. Post-factum applies even more

materially to measuring the performance of

money managers, and in that regard it is not a

trivial matter.

The next one on the list, the self-ful½lling

prophecy, or 

sfp

as he called it: Some econo-

mists have speculated that the option pricing

model that Fischer Black, Myron Scholes, and

I derived might not have been descriptive of

market pricing before its public release, but if

enough people believed in it, maybe prices ad-

justed so the model became accurate afterwards.

Then there was 

rkm

’s concept of manifest and

latent functions, well cited in my early papers

but even more so in my later work applying

functional analysis to better understand ½nan-

cial innovation and institutional change. Con-

sider a speculative market, say the stock mar-

ket. Its manifest function is to permit transac-

tions that allow ½rms to raise funds by issuing

shares that investors can buy and trade amongst

each other. So if you were never going to do any

transactions in the market, why then would you

have any interest in it? One answer is that the

market gives you information. Managers who

might never issue equity in their ½rms never-

theless need to look to the market for infor-

mation to help them make effective decisions

about the ½rm’s investments–thus providing

information is an important latent function of

speculative markets.

Still another one of

rkm

’s formulations, the

concept of unanticipated and unintended con-

sequences of social actions, came into my work

when I was analyzing the implications of the

U.S. government providing guarantees for cor-

porate pensions. In the last decade I’ve focused

on developing the so-called functional perspec-

tive in ½nance to examine the evolving ½nan-

cial system. This perspective uses ½nancial

functions instead of institutions as “anchors,”

or framing elements of the system–thus free-

ing us to think about the dynamics of change

in economic institutions as an endogenous pro-

cess. You can imagine the discussions I had with

my father about that. So I ½nd myself very, very

close to precisely where I was trying not to be

at the start, more than four decades ago–and

darn happy about it! 

Now, apart from our endless talks, my father

and I did actually work together: in 1981 we

had a draft of a theory including a mathemati-

cal model on problem choice and the functions

and dysfunctions of priority in the reward sys-

tem in science. Although subsequently both of

us became very involved in our other separate

research projects, we continued to work to-

gether to expand on the ideas and potential ap-

plications. In fact, in 1989, our joint paper was

going to be the lead article in Rationality and So-

ciety, the journal that James Coleman had just

started, but my father did what he had done

more than once before. As it was about to go to

press he said, “No, it’s not quite good enough.”

Fifteen years later, it’s still not good enough,

but I’m going to try to make it right. In the fall

of 2002, my father published his last book, The

Travels and Adventures of Serendipity (coauthored

with Elinor Barber), which actually originated

in the 1950s. So, in terms of his holding back

and delaying publication, our joint project is

nowhere close to the record.

In sum, try as I could to do otherwise, I ended

up not at all far from my father. He was indeed

a role model in the deepest sense of that term.

He was so aware and so active and so intellec-

tually productive up to the end of his life–and

that’s a heck of a role model to have going for-

ward.

I am really delighted that this meeting is taking

place here. This was a very important institu-

tion in my father’s life, and I had the great good

fortune to join the Academy and be a part of it

while he was too. Indeed, we had several such

shared connections, including membership in

the National Academy, and we were the only

father and son ever to hold honorary degrees

from the University of Chicago.

I cannot end my remarks without noting an-

other collaboration that never occurred–the

one in which the two of us coauthor a piece

with Robert Merton Solow. My father said to

me, “We simply must do a joint piece with Bob.”

And I immediately reacted, “That’s great . . .

let’s do it! I surely wouldn’t mind being sand-

wiched between the two of you.” He elaborat-

ed, “Of course, there can be only one way to

order the authors–Robert Merton Solow and

Robert Merton Duo.”

Robert M. Solow

I

am not a sociologist, as you know, nor have I

ever taught at Columbia University. So far as I

can recall, in fact, I have never been formally

or institutionally associated with Bob Merton,

senior. There is, however, one thread of con-

nection between Robert K. Merton and Robert

Merton Solow that goes back more than sixty

years–to January 11, 1942, to be precise. (There

is someone here this evening who goes back

even further: Ruth Smullin, the wife of my

mit

colleague Louis Smullin, was Merton’s

tutee as a Radcliffe student in 1938.)

If you look at the bibliography in the 1970 re-

print of Merton’s 1938 classic Science, Technology

and Society in Seventeenth-Century England, you

will ½nd a reference to a paper by one Robert

Solow entitled, of course, “Merton’s Science,

Technology and Society in Seventeenth-Century En-

gland.” The paper is dated January 11, 1942. On

January 11, 1942, I was a sophomore at Harvard

College, a little less than seventeen and a half

years old. I had been taking Talcott Parsons’s

course on the sociology of science, and had

turned in my term paper on time. Parsons gave

it an A and sent it on to Merton. With his usu-

al grace, Bob Merton read it, entered it in his

bibliography, and found it unnecessary to com-

ment on it. I live in hope that the paper is irre-

trievably lost or destroyed. I am afraid I can

imagine what it was like. I didn’t get to know

Merton personally until much later. 

There is no mystery in the fact that he was un-

appeasably attracted to all kinds of ideas, espe-

cially ideas about social institutions, but really

about anything. What is more mysterious is

that somehow ideas were attracted to him, as

if he were some kind of intellectual flypaper. If

we took any six of us here and put us in a room

with Bob Merton and released an idea some-

where near the chandelier, the odds are two to

one that it would flutter down and come to rest

on Merton. You realize what that means: the

random idea had a two-thirds probability of

½nding its way to Merton and a probability of

My research is very mathe-

matical, very quantitative,

and seemingly very far away

from what my father did.

Nevertheless I found myself

using many of his ideas.

---

18 Bulletin of the American Academy   Spring 2004

and that comes as no surprise. He cannot have

had a monopoly on sanity, but he had a healthy

market share; it was clearly his natural tem-

perament.

I want to illustrate this discussion of the inter-

connection issue by an anecdote. Bob Merton

was not a participant, but I dearly wish he had

been. To begin with, in the course of the mid-

dle-range essay he mentions that Durkheim’s

Suicide is his nominee for the Academy Award

for middle-range theory, with Weber’s The

Protestant Ethic and the Spirit of Capitalism maybe

a close second. (The Academy Award locution

is mine, not his. Like many of you, I have read

both books, but missed the movie versions.) 

Now the story continues. Back in 1957–1958,

by chance, Talcott Parsons, David Landes, and

I were all fellows at the Center for Advanced

Study in the Behavioral Sciences in Palo Alto.

One afternoon, Landes was giving the weekly

seminar. He had (serendipitously) come upon

a treasure trove of personal letters written by

members of some entrepreneurial textile-

manufacturing families in the cities of Roubaix

and Turcoing in northern France. The letters

included hortatory messages from fathers and

uncles to sons and nephews. These people were

all pious, unquestioning Catholics. But the

messages they were transmitting and the ad-

vice they were giving sounded exactly as if

they had been lifted from Weber’s Calvinists.

So what is then to be made of the Protestant

ethic and the spirit of capitalism? Could We-

ber’s middle-range theory have been wrong?

It was in Bob Merton’s mind that one of the

great advantages of middle-range theories is

that they could be wrong and you could show

them to be wrong. Talcott Parsons, who sort 

of owned Max Weber in those days, would not

allow the possibility of error. Weber hadn’t

actually meant Protestantism, but maybe just

some appropriate religious orientation, or even

some ethically sanctioned canons of behavior

that could in principle originate outside of for-

mal religion altogether. 

What would Merton have said had he been

there? I think he might have said: “OK, so it’s

not exactly Protestantism, and Weber erred.

But there is a research project here. Where,

across time and space, have there arisen simi-

lar ethical injunctions to achieve demonstrat-

ed goodness by worldly success? And where

have they been broadly accepted? And what

do those instances have in common? Are there

other well-documented middle-range theories

that bear obliquely on this sort of thing?” This

would, of course, have been put elegantly, and

with a reminder that to proceed in that direc-

of humor.) It is clear that Merton does not

think it is fruitful to start by deducing middle-

range theories from such comprehensive theo-

retical systems. In the ½rst place, all those grand

theories in sociology are more like “theoreti-

cal orientations” than they are like tight, rigor-

ous theories. So it may not be possible to get

from some grand theory to a particular middle-

range theory. Merton points out explicitly that

a successful middle-range theory may be logi-

cally compatible with more than one grand

theory. I don’t remember that he says so explic-

itly, but it seems to me that, in sociology, one

grand theory may be compatible with more than

one middle-range theory aimed at the same

group of observable facts. But it is pretty clearly

Merton’s view that if a successful–that means

empirically successful–middle-range theory

is incompatible with some overarching system,

so much the worse for that grand system.

It is very important for him that middle-range

theories have the potential to interconnect with

one another and form networks of related hy-

potheses that can give rise to broader general-

izations; but the examples of this that he men-

tions, without developing them, seem to in-

volve fairly small increments in generality. It is

not clear to me whether Merton harbored the

thought that very general social theory could

be built from the bottom up by this process of

accretion. But he made no bones about where

he thought the opportunities for productive

work in sociology and social theory were to be

found–and my experience in economics leads

me to believe that he was right. 

Not every sociologist agreed with him. The

paper gives some examples of polarized argu-

ments pro and con the middle range. I am not

knowledgeable enough about what went on

before and after to try for a balanced interpre-

tation. But I will say that Merton’s discussion

of the controversy is distinguished by sanity,

just one-eighteenth of coming to rest on one of

us. And we are intelligent, idea-friendly people

or we wouldn’t be here. He must have emitted

some pheromone-like come-on.

It struck me that the obviously right way for me

to pay tribute to my friend–and everybody’s

mentor–is to stick to ideas. So I reread one of

Bob Merton’s most celebrated papers, and I

propose to describe it and discuss it now. The

paper I have in mind is “On Sociological Theo-

ries of the Middle Range.” I chose it partly for

sel½sh reasons: I think of myself as an econom-

ic theorist of the middle range, so I am curious

about parallels and differences between the

disciplines.

Merton tells us that this essay expands some

comments he had made on a paper read by Tal-

cott Parsons at the 1947 meeting of the Ameri-

can Sociological Society. Merton’s essay can be

read as a sort of declaration of independence

from his teacher and friend, because Parsons

was then the leading advocate and exemplar of

the tendency in social theory that Merton was

opposing. He says that serious work in sociolo-

gy ranges from single, isolated empirical stud-

ies of particular situations with little or no po-

tential for generalization or even for intercon-

nection, all the way to grand, all-inclusive sys-

tems of sociological theory. He thinks that the

best prospects for progress in sociology lie in

theories of the middle range: “theories that lie

between the minor but necessary working hy-

potheses that evolve in abundance during day-

to-day research and the all-inclusive systematic

efforts to develop a uni½ed theory that will ex-

plain all the observed uniformities of social be-

havior, social organization and social change.”

And later: “Middle-range theories consist of

limited sets of assumptions from which spe-

ci½c hypotheses are logically derived and con-

½rmed by empirical investigation . . . . These

theories do not remain separate but are con-

solidated into wider networks of theory.”

Merton offers, as sociological examples of the

middle range, the theory of reference groups,

the theory of social mobility, of role conflict,

and of the formation of social norms; and he

thinks of these as analogous to such classical

instances as a theory of prices, a germ theory

of disease, or a kinetic theory of gases. 

A dif½cult question arises about the relation of

middle-range theories to total systems of soci-

ological theory, like those associated with such

names as–I have to quote this–“Marx and

Spencer, Simmel, Sorokin and Parsons.” (It was

certainly not lost on Merton when he wrote

down that passage that neither Karl Marx nor

Herbert Spencer had even a microscopic sense

There is no mystery in the

fact that he was unappeas-

ably attracted to all kinds

of ideas . . . . What  is  more

mysterious is that somehow

ideas were attracted to him,

as if he were some kind of

intellectual flypaper.

---

Bulletin of the American Academy   Spring 2004    19

tion would be to stand on the shoulders of Max

Weber rather than to sit at his feet.

The spirit of Merton’s stern but reasonable

essay went down so smoothly with me that I

naturally began to think about its relevance for

economics. It seems to me that there are two

characteristics of economics as a discipline that

matter in this respect. The ½rst is that the very

objects that economic theorists think about

are very narrowly de½ned: economic theory is

about explaining and understanding prices and

quantities of goods and services. If what you

are saying is not at least indirectly about prices

and quantities, then you are not talking about

economics. I don’t want to ask whether this

restriction is somehow inevitable, or whether

there could be a recognizable economics that

was about something else. The fact is that prices

and quantities are what essentially all econom-

ics is about.

This fact–historical or logical or accidental–

creates two contrasts with sociology. The ½rst

is that it limits drastically one’s freedom to

create grand theories in economics. Grand

theories often differ from other ones because

they change the subject, they introduce wholly

new, or apparently new, concepts. If it all has

to come down to the price of orange juice in

the end, you can try to go at the problem differ-

ently, but not too differently.

The other contrast is that prices and quantities

are numerical. There is no getting away from

the fact that economics has to be quantitative.

I don’t think that the use of mathematics is all-

important, though it certainly matters for the

day-to-day business of working economists. It

is fundamentally the more-or-lessness that fo-

cuses economists on the middle range.

Merton had thought about this too. He speaks

of the virtue of “bring[ing] out into the open

the array of assumptions, concepts, and basic

propositions employed in a sociological analy-

sis. They thus reduce the inadvertent tendency

to hide the hard core of analysis behind a veil

of random, though possibly illuminating com-

ments and thoughts . . . . [S]ociology still has

few formulae–that is, highly abbreviated sym-

bolic expressions of relationships between so-

ciological variables. Consequently, sociological

interpretations tend to be discursive. The logic

of procedure, the key concepts, and the rela-

tionships between them often become lost in

an avalanche of words.” He suggests that one

of the important functions of mathematical

symbols is to provide for the simultaneous in-

spection of all terms entering an analysis.

All this has led to a situation in economics that

would have amused Bob Merton. I think it has

to do with the focusing role of prices and quan-

tities. Only one overarching system, or para-

digm, survives: it is what is uninformatively

and vaguely called “neoclassical economics.”

As an inclusive theory it has serious problems

of its own. Generally, however, even important

subversive movements seem rather to modify

it but not to replace it, although some of its

characteristic conclusions have to be given 

up. I will mention two outstanding examples:

Keynesian economics gives up on the assump-

tion that prices and wages move flexibly and

quickly to equate supply and demand for labor

and goods; this is a big deal, but leaves the basic

paradigm still quite recognizable. Secondly, so-

called behavioral economics wants to give up,

in some circumstances, the clearly over-simple

assumption that economic decision making is

governed exclusively by greed and rationality.

It is not so clear what to put in its place, how-

ever. Recent research on the role of reciprocity

in economic behavior would have fascinated

Merton; he would also have noticed that this

is middle-range theory of the best kind. In any

case, the ½ndings of behavioral economics

would change some of the implications of the

paradigm, but not fundamentally its basic out-

line. 

Is this resiliency a good thing? Too much re-

siliency eases over into flabbiness. I didn’t like

it much when Parsons tried to defend Weber

against Landes’s observations. But that was 

a middle-range theory, not a “paradigm.” A

middle-range theory is supposed to live or die

by its empirical bite. Neoclassical economics

is, in Merton’s phrase, a “theoretical orienta-

tion.” This is another case where I can imagine

my half of a conversation with Merton, but I

am not so sure about his.

All that is interesting but not amusing. What

would have brought a smile to Bob Merton’s

face is the fact that defenders of the pure, un-

adulterated neoclassical system have found

themselves forced by the threat of Keynesian

subversion to reduce the all-inclusive system

to a single middle-range sort of model and then

try applying this to the data of everyday macro-

economic life, where it has, on the whole, had

no real success. Merton, thou shouldst be with

us at this hour.

I hope I have said enough to show you that in-

tellectual engagement with Bob Merton was

always an exhilarating experience for a social

scientist, even for an economist. He was right

about the useful scope for social theory, and he

had an unerring eye for intellectual puffery and

pomposity. Sadly for me, I only got to know

him well in his later years; and this leaves me

uncertain about what he was like in his forties

and ½fties, at Columbia and, with Paul Lazars-

feld, at the Bureau of Applied Social Research.

I had intended to leave aside the celebrated 

On the Shoulders of Giants and the soon-to-be-

celebrated Travels and Adventures of Serendipity.

After all, they are just books about the origins

of particular words and phrases and how they

acquire and change their meanings as they are

transmitted from person to person and place

to place and time to time. But I decided at the

last minute that I couldn’t just forget them:

those books say too much about the Merton I

knew. 

They are, of course, a lesson in how to wear

learning gracefully. Maybe it is a lesson we don’t

need, because so few of us have that much learn-

ing to wear. They also demonstrate a much more

complex quality. They tell their story. Then

they think about the social relations of the peo-

ple who populate and carry the story, mostly

scientists and other intellectuals. Then they

reflect on the act of doing the sociology of sci-

ence and the sociology of knowledge. And then

somehow they take an unbemused but toler-

ant view of the whole act of writing the book.

There may be further layers that I am not clever

enough to detect. One does ½rst think of Tris-

tram Shandy. But Sterne is artless compared

with Merton. No, that’s not fair: Merton knew

about Tristram Shandy, stood on his shoulders.

What I wonder about in particular is how much

of the later Merton persona was already visible

in those earlier years. By the time I knew him,

his breadth of knowledge and depth of pene-

tration were lightly covered over by good na-

ture. But no one after ½ve minutes would have

mistaken him for a harmless old codger. He

was in fact the embodiment in the intellectual

sphere of Muhammad Ali’s description of his

By the time I knew him, his

breadth of knowledge and

depth of penetration were

lightly covered over by good

nature. But no one after ½ve

minutes would have mis-

taken him for a harmless

old codger.

---

20 Bulletin of the American Academy   Spring 2004

it. He used to say that he would sometime like

to write a joint paper with his son Bob and with

me. He wasn’t sure about the topic, but it should

be signed Robert Merton Duo and Robert Mer-

ton Solow. Two thirds of the authorship of that

unwritten paper are here, but the best third is

missing.

© 2003 by Robert C. Merton and Robert M.

Solow, respectively.

own style: Float like a butterfly, sting like a bee.

I believe ½rmly that anything worth doing is

worth doing light-heartedly; and I could point

to Bob Merton as an adequate demonstration. 

I said that I had never been formally associated

with him. That is true. But I did overlap with

his wife Harriet for some years on the board of

the Center for Advanced Study in the Behavior-

al Sciences. Sharing breakfasts with Bob and

Harriet in the hotel dining room before those

meetings taught me what I wanted to be when

I grew up. I don’t even care that I never made

2

1.

Robert C. Merton and Robert M.

Solow

3

4

5

1

2.

Harriet Zuckerman (Andrew W.

Mellon Foundation) and Academy

President Patricia Meyer Spacks

3

.

Barbara Rosenkrantz (Harvard Uni-

versity) and Werner Sollors (Harvard

University)

4

.

Victor Brudney (Harvard Law School),

Patricia Albjerg Graham (Harvard

University), and Loren Graham (MIT)

5

.

Visiting Scholars Jonathan Hansen

and Adam Webb