The Patent Office
At the Swiss Patent Office, Einstein’s job was to consult with
inventors. Only
one of his written opinions on patent applications is left , the
administrative
documents having been routinely destroyed. This opinion is on an
alternatingcurrent
machine with short- circuit brushes and opposing auxiliary spools for
spark prevention. Einstein off ers not a single good word for the patent
claim:
it is “incorrect, inaccurate, and unclear.” He also gave the
requirements for a
correct application: it should note only characteristics of the subject
of the
patent that are in the claim, and each par tic u lar embodiment should
correspond
to the main claim of the main patent and the claims of the actual
patent.
? His advice did not help the inventor. A revised application was
submitted
two months later, but Einstein was still not satisfi ed with it.
We have indirect evidence of a further case. A certain Ignacy Moscicki,
inventor
of a way to produce nitrogen acid from the atmosphere, of a new method
of concentrating nitrogen acid and sulfuric acid, and of high power
capacitors,
submitted an application to the Swiss Patent Offi ce in 1906. It dealt
with an arc furnace for the production of nitric acid, in which the arc
was
rotated by an electromagnet.? As the designated expert, Einstein was
espechapter
three
Expert Opinions
38 The Practical Einstein
cially interested in why the electric arc changed its orientation in a
magnetic
fi eld. He gave a positive opinion of the application. The story, fi rst
aired in
1934,? was rediscovered by Zofi a Golab-Meyer.?
That Moscicki and Einstein did meet in Bern is confi rmed by an exchange
of letters in 1932. Einstein asked Moscicki for help in getting a
position for an
acquaintance of his. In the introductory sentences, he remarked that “I
know
that you were originally a physical chemist, and I hear that even now
you work
as an or ga niz er for scientifi c and technological research.” ? The
letter was addressed
“To Mr. President of Poland Professor Dr. Moscicki.” Moscicki, the
successful inventor and scientist, returned to Poland in 1912, where he
was
named professor of chemical physics and technical electrochemistry at
the
Technical University of Lwów (present- day Lviv, Ukraine), and in 1926
was
elected president of Poland.
In reply, Moscicki emphasized the meager prospects in Poland for
Einstein’s
protégé, but he also mentioned that Einstein’s letter gave him par tic u
lar pleasure,
because it reminded him of their meeting in Bern and later in Fribourg.?
From Einstein’s next letter we learn that he remembered their meeting
with great satisfaction, especially the one in Fribourg with Kowalski.?
The
encounter in Bern may have been in the patent offi ce, but the
University of
Bern cannot be excluded either, for Einstein succeeded in getting the
venia
legendi (permission to teach) there around February 28, 1908, and he was
Privatdozent
there from April 21, 1908, to August 4, 1909. In May and June 1908, he
also worked at the University of Fribourg in Professor Albert Gockel’s
laboratory
on his fi rst invention, an electrometer for small quantities of
electricity
(see chapter 4), and on May 24, 1909, he attended a physics colloquium
there.?
In addition, Professor Joseph Kowalski (Józef Wierusz- Kowalski),
professor
of physics at the university, was interested in this electrometer,?? so
there was
ample opportunity to meet Moscicki, for he served as Assistent to
Kowalski
from 1896 to 1912.
Three more patents are supposed to have gone through Einstein’s hands:??
an electrical typewriter with shuttle- type carrier,?? a gravel
sorter,?? and a meteorological
station controlled by ambient humidity.??
Gyrocompasses
Hermann Anschütz- Kaempfe, the inventor of the gyrocompass and own er of
a gyrocompass factory in Kiel, Germany, sued the American Sperry
Gyroscopic
Expert Opinions 39
Company.?? The competition between them had turned fi erce when, in
1914,
Sperry sold a compass to the German navy. Selling gyrocompasses was a
promising
business at the time. They could be used on ships, submarines, and
airplanes,
because metal structure does not distort their indication as it does
with magnetic compasses. In addition, in the de cade before the First
World
War, Germany intended to build a navy comparable to or outrivaling
Britain’s.
One of the Anschütz patents allegedly infringed was DE182855,?? which
protected the original design. Sperry claimed that the patent was void,
for it
was not new when compared to a previous patent of Marinus G. van den
Bos.??
The other patent that Anschütz claimed to be infringed, DE236200,??
described
a means of damping unwanted oscillations.
The fi rst hearing of November 10, 1914, was adjourned, but the court
advised the parties to choose an impartial expert living not far from
Berlin
in order to keep expenses low. The court submitted a list of potential
experts,
including Arnold Sommerfeld, professor in Munich, and Felix Klein,
professor
in Göttingen, who were coauthors of a book on gyroscopes.?? The fi rst
name
on the list was, however, Einstein’s, perhaps because he was a Berliner,
and the
court selected him. A court expert’s duty was to answer questions
impartially,
to attend the proceedings, and to provide oral testimony.??
The next hearing took place on January 5, 1915. Einstein failed to
impress
the court; he was not well prepared. To make his task easier, the court
formulated
four questions to be answered in a written report and presented at the
next hearing:
1. What are the physical principles of gyrocompasses?
2. What are the diff erences between Anschütz’s compass and other
gyrocompasses, in par tic u lar the compass patented by Van den Bos as
DE34513?
3. What is the gist of the invention patented in DE236200; and did this
invention make it possible to produce the fi rst, perfectly working
gyrocompass?
4. What are the similarities and diff erences between the Anschütz and
Sperry compasses, and had Sperry made use of Anschütz’s two
inventions to such an extent that his compass was technically similar
to Anschütz’s compass?
Einstein answered the questions on February 6, 1915.?? He gave a clear
exposition
of the principles of gyroscopes and gyrocompasses (question 1) and
40 The Practical Einstein
declared that Anschütz was the fi rst to produce the fi rst usable
gyrocompass
with damped oscillations (question 3), which he achieved in his second
patent
with a method better than Van den Bos’s. Einstein denied the novelty of
Anschütz’s fi rst patent, giving priority to Van den Bos (question 2).
Finally, he
declared that Sperry did make use of Anschütz’s fi rst patent but
doubted that
the second was infringed (question 4), that is, he accepted Sperry’s
damping
as an original idea.
Apparently neither the court nor Anschütz was happy with Einstein’s
voting
for Sperry. In the second hearing on March 26, Einstein did not convince
the court with his opinion, so he was given two further questions.
The fi rst question practically repeated the previous question 2, asking
him
to explain the relationship between Anschütz’s and Van den Bos’s
patents.
The second question reformulated the earlier question 4: How far had
Sperry
made use of the ideas in Anschütz’s two patents in his own compass
delivered
to the German navy?
Einstein prepared a supplementary opinion on August 7.??
Aft er a meticulous study of the patent specifi cations, he concluded
that
Anschütz’s fi rst patent did indeed diff er from Van den Bos’s
invention.
To answer the second question, Einstein tested a Sperry compass. He
confi rmed that Sperry made use of Anschütz’s fi rst patent, but, in
contrast
to his fi rst opinion, he declared that the same is true for the second
patent.
This time he received no further questions, and the court decided in
favor
of Anschütz.
Why did Einstein change his mind? Apparently he had not immersed himself
in the case deeply enough. When in 1918 Anschütz asked him to serve as
his private expert in another proceeding, Einstein expressed the self-
critical
hope that in the future “the insuffi cient understanding of impartial
experts”
would not cause damage to Anschütz and make him angry.??
In this case, Anschütz- Kaempfe considered that a patent application of
the
Gesellschaft für nautische Instrumente (GNI)?? infringed upon his patent
DE241637.?? The GNI invention was an arrangement to avoid erroneous
indication
of gyrocompasses when the ship is rolling. He asked Einstein to serve
as a private expert on his behalf?? and solicited an opinion from him on
whether
the method proposed by the GNI fell within the scope of protection of
his patent
or not.?? Einstein’s opinion submitted on July 7, 1918, is not
available. Apparently
Anschütz was not satisfi ed with it, and he requested another one.??
Expert Opinions 41
Einstein began his second opinion with an analysis of how the motion of
a ship infl uences the gyroscope (fi g. 3.1).??
Let the rotating gyroscope be suspended from P by a solid rod a. If it
swings
between A" and A' in a plane to which its axis of rotation stays
perpendicular,
it will not change its direction. But if it swings perpendicular to the
previous
plane (between B" and B'), its axis of rotation will oscillate
perpendicular to
the plane of its swing, but the time average is zero. Consequently, the
direction
change will not present an insurmountable diffi culty in indicating the
correct cruising direction.
If, however, the gyro swings in a direction that is a combination of
these
two directions, a torque may appear that makes the gyroscope’s axis of
rotation
rotate with respect to the vertical. Both the Anschütz and the GNI
inventions
aim to eliminate this eff ect.
Anschütz’s patent gives two ways to avoid or minimize this eff ect: to
prevent swings from A" to A' (which is eff ectively the same as allowing
only
swings from B" and B'), or to use two or more gyroscopes with their axes
not
parallel to each other. If two gyros are mounted in a frame this way,
any rotation
Figure 3.1. Gyroscope motion.
42 The Practical Einstein
of the axes is prevented by inertial forces if the angle between the
axes is kept
fi xed by a nonrigid connection. The patent does not stipulate that the
gyroscopes
be horizontally arranged; the only requirements are that the resulting
moment of all the gyros has a horizontal component on which gravitation
acts
and that the gyro axes be nonrigidly connected and not positioned
parallel to
each other. Einstein declared that the claims are expounded so clearly
that, by
following them, any engineer with a knowledge of the subject can build a
usable
gyrocompass.
Then he turned to the GNI patent application. It also uses two
gyroscopes
with nonparallel axes and connected with a nonrigid connection, so it is
evident
that it falls within the main claim of the Anschütz patent. The only
question
left is whether its construction represents a technical improvement.
Einstein said no.
He concluded that the subject of the GNI application fell within the
scope
of protection of the Anschütz patent, and its specifi c features are
neither novel
inventions nor practical developments.
Maybe the applicant succeeded in presenting arguments to prove that
these specifi c features did represent novel invention; maybe the
lawsuit took a
diff erent course for other reasons. We do not know, because the
documents of
the court and patent administration are no longer available. We only
know that
Anschütz lost the case and that a patent,?? along with two additional
patents,??
was granted to GNI in 1918.
The encounter was, however, not settled at this point. In the spring of
1922,
the manager of GNI, Professor Oscar Martienssen, approached Einstein and
requested that he withdraw his opinion of 1918 because Anschütz intended
to
use it against GNI.?? He argued that Einstein’s opinion was based on
errors,
and he wanted to avoid protracted and time- consuming discussions in
various
courts that would take Einstein’s “invaluable talent” away from “more
important
things.” If Einstein agreed, there would be no need of attacking him in
court. From the formulation, one may infer that Anschütz had lodged
another
suit against GNI.
Martienssen called Einstein’s attention to a mistake that he himself had
made in an earlier publication,?? but later corrected.?? He also added
that Richard
Grammel followed the same erroneous considerations in his book on
gyroscopes,
?? but later he also realized it was a mistake. Apparently Martienssen
supposed that Einstein had relied on these publications when he prepared
his
opinion.
Expert Opinions 43
Martienssen’s second objection was that Einstein had given no reason why
the patent application was dependent on Anschütz’s patent. The Anschütz
patent states explicitly that the invention refers to a supplementary
gyro in the
moving system and not to a system with a gyro for stabilizing the
cardanic
suspension. The GNI application, however, uses a stabilizing gyro.
Furthermore,
Anschütz’s patent can be constructed using two auxiliary gyros but not
with only one, with its axis perpendicular to the northward directing
gyro.
The spring that joins the two gyros is of fundamental importance for the
Anschütz patent, whereas for the GNI application the vertical gyro is fi
rmly
mounted on the ground plate.
“You may rest assured,” Martienssen added, “that I feel terrible about
writing
these lines to you whom I sincerely hold in high esteem for your eminent
achievements.”
Einstein did not understand the letter, simply because he did not
remember
the proceedings that had taken place four years earlier.?? He asked
Martienssen
for clarifi cation, forwarded Martienssen’s letter to Anschütz the same
day,??
and doubted that Martienssen was right, “despite his arrogant tone.” The
following day, Martienssen replied to him and attached copies of the
relevant
patents,?? as well as a copy of Einstein’s 1918 opinion. He also added a
further
objection: in his opinion, Einstein had explicitly stated that
Anschütz’s
direction- indicating system was equipped with two or more gyroscopes
and
that in the patent of GNI the same system has only one gyro. Why was it
then
that in his further considerations Einstein mentioned two gyros in the
GNI
patent application? In it, the stabilizing gyro has nothing to do with
the direction
indication.
The hearing began on April 11, 1922, in Kiel.?? On April 12 Einstein
prepared
a supplementary opinion, but it is not available. Anschütz won the
case.??
GNI submitted an appeal to the Higher Regional Court on June 9.
Einstein prepared a second supplementary opinion.?? In the particulars
of
the appeal, he wrote, a patent was mentioned to demonstrate that there
were
patents on gyrocompasses with constructions similar to Anschütz’s
invention
but had not been considered to infringe upon it.?? This patent did not
include
means to eliminate or minimize the rolling error. The opposite
statements in
the particulars are untrue.
Then he refl ected on the opinions of two experts. The fi rst maintained
that
the GNI application had a specifi c feature that Anschütz’s patent does
not.
Einstein replied that the same feature was explicitly described in it
and that,
44 The Practical Einstein
at present, it is not the complete in de pen dence of the patents at
stake but the
question of whether the GNI patent is dependent on Anschütz’s patent or
not.
He summarized his reply to the opinion of the second expert in three
points:
1. Anschütz’s patent is the fi rst to realize that the rolling error
depends
on the swing period of the directional system around the gyro axis,
and it is the fi rst to off er a means to avoid or signifi cantly reduce
this
error.
2. The inspected instruments (apparently each fi rm presented one of its
own) use gyroscopes to reduce swinging.
3. In these instruments, the gyros that slow down the swing are not
mounted directly on the base of the directional gyro but on a component
part, which is connected to the others with rigid connection.
Anschütz’s patent would be already infringed if only the fi rst point
were a
feature of the GNI patent, but because all three points are shared by
it, the
GNI patent is technically the same as the Anschütz patent.
There was a second session on July 10, 1922. Einstein’s main role was as
a “bogeyman.” ?? Anschütz won again. The last information on this suit is
that
“the scoundrel did not get away with his tricks” ??— an impolite and
unjust
remark from Einstein’s pen for, as we saw, Martienssen was polite and
raised
clear technical objections to Einstein’s opinion. While Einstein’s fi
rst reaction
to Martienssen’s approach was to call his tone “arrogant,” we have ample
evidence
for Schell’s remark, that “the impartial expert had long since become a
good and thoroughly partial friend of Anschütz- Kaempfe and his fi rm.”
??
In Einstein’s next case, the defendant was Franz Drexler, and again the
plaintiff Anschütz- Kaempfe.?? Drexler, a trained pi lot, had been
working on
a gyrocompass with Anschütz and, aft er having left the company, tried
to
sell it through his newly founded company, the Kreiselbau Co. The
invention
was a gyrocompass for indication of vertical and horizontal turns of
airplanes.??
In his opinion of July 23, 1919, Einstein set out by explaining the
behavior
of a gyroscope that has two degrees of freedom (fi g. 3.2).??
Let a gyroscope be mounted in an inner gimbal R, which can turn around
axis B–B in an outer gimbal G. Springs F restore the position of R
whenever it
is forced to leave the plane of G. Let the gyroscope be mounted on an
airplane
with the plane of G parallel to the wings. When the airplane gains or
loses
Expert Opinions 45
height, R will not leave the plane of G, because its axis of rotation
A–A is
shift ed parallel. When, however, the airplane turns to the right or
left , R will
step out of this plane by turning around B–B and will keep the new
position
until the turn is fi nished. Einstein calls this arrangement the “turn
indicator.”
How can up and down turns be indicated? Maybe with a simple plumb line.
Let us suspend such a line L in a plane P and keep P horizontal (fi g.
3.3). Until
they move uniformly, L and P will make a rectangle. If P is accelerated,
L will
lag behind, and so they will not stay perpendicular to each other. The
same
happens should P turn up or down.
This simple pendulum cannot provide a precise indication of up and down
turns, for it is possible that if the airplane is accelerated and turned
downward
at the same time, L and P could remain perpendicular. Let us call this
device,
as Einstein did, the “plumb indicator.”
Now return to the gyrocompass and fi x R on G in a line lying higher
than
the center of mass of R and K. In addition, remove the springs F. By
doing so,
we combine a gyroscope and a plumb indicator, which, however, cannot
indicate
directly whether the airplane is turning or not. Einstein calls this a
“gyro
pendulum.”
Aft er these preparatory considerations, Einstein answered fi ve
questions
put by the court. Because we do not have the original court document,
and
Einstein’s answers do not follow the numbering of the questions, I can
only
summarize them.
Figure 3.2. Gyrocompass.
Albert Einstein, “Court Expert
Opinion in the Matter of
Anschütz & Co. vs. Kreiselbau
Co.,” July 23, 1919. Courtesy
Albert Einstein Archives, The
Hebrew University of Jerusalem.
46 The Practical Einstein
Anschütz’s patent consists of a gyro pendulum and a plumb indicator.??
The
indications of the two instruments must be compared to follow the fl
ight on a
general curve. Even though this is not the fi rst patent that indicates
the change
of orientation,?? it is the fi rst to indicate vertical turns.
Anschütz can claim priority not in the application of a gyro pendulum
but
in the combination of this pendulum with the plumb indicator, and,
because
it is only this combination that can help fl ight in a curve, the patent
represents
technical progress of inventive importance.
Drexler’s patent follows the same principle as Anschütz’s, Einstein
continued:
it makes use of two gyroscopes with horizontal axes (even though
arranged
as a turn indicator) and of a plumb indicator. Whether the plumb
indicator
is a separate device, as with Anschütz’s, or it is the turn indicator
itself
that is turned into a plumb indicator as with Drexler’s, makes no diff
erence
of principle. There is, however, a technical diff erence between them:
Drexler’s
device indicates right and left turns directly, not as a diff erence
between the
indications of two instruments.
Einstein felt that this opinion sounded obscure for the lawyers, so he
appended an explanation, as he put it, in a “freer form,” an application
of
set theory.
Figure 3.3. Turn indicator.
Albert Einstein, “Court Expert Opinion in the Matter of Anschütz & Co.
vs.
Kreiselbau Co.,” July 23, 1919. Courtesy Albert Einstein Archives, The
Hebrew University of Jerusalem.
Expert Opinions 47
Consider a plane P whose points represent all the possible technical
realizations
of all the patent inventions. The embodiments of a par tic u lar
invention
make a region G of this plane. Had the inventor a complete knowledge of
all the embodiments of his invention, that is, of G, he should be
considered as
the only own er of them. He has, however, only a limited knowledge of G;
there
can be various technical embodiments of his idea of which he does not
know,
and which may have novel technical features. Such cases can be called
“dependent
inventions.” Whether such an invention may have some legal rights
is a problem for lawyers.
Drexler’s patent is a “dependent invention” in this sense: dependent on
Anschütz’s patent, for it serves the same goal and uses the same
gyroscopes,
but it is a genuine invention, too, for it uses only one indicator in
place of two,
and so it produces its result more safely and with greater precision.
Drexler’s
patent covers Anschütz’s patent, but it does not copy or circumvent it.
Neither parties were satisfi ed with Einstein’s opinion. According to
Anschütz’s patent lawyer, Hugo Licht, it was “not negligibly weak”;??
therefore,
Einstein was requested to fi nd new arguments in Anschütz’s favor.
Einstein called the director of regional court and explained that when
he had formulated his opinion, he was staying in Switzerland and, upon
returning to Berlin, he found documents that he had not been able to
use.
He would like to complete his opinion by taking them into
consideration.??
He submitted a supplementary opinion on October 9, 1919, of which only
one
paragraph is extant, quoted in Licht’s letter to Anschütz.?? In it, he
declares
that it was Anschütz’s patent that fi rst proposed a gyroscope with two
degrees
of freedom and with a horizontal axis of rotation to keep track of the
direction change of the aircraft , and Drexler’s turn indicator is also
based
on this idea.
On the hearing of November 4, under the pressing questions of Drexler,
Einstein admitted that if a pi lot had known of a paper published in
1910 on
a turn indicator,?? it would not have been necessary to wait for an
invention.
With this, he weakened the technical importance and priority of
Anschütz’s
1917 patent. Licht explained Einstein’s point by saying that an eminent
scientist
uses stronger criteria than a judge for what can be considered an
invention.??
Anschütz won the case, but Drexler appealed. Einstein was again proposed
as an expert, but because the court expected a second expert who was
“well
48 The Practical Einstein
informed in both the theory of gyroscopes and the behavior of airplane
during curved fl ight,”?? Anschütz proposed two further candidates,
Richard
Grammel, professor at the Technical University of Stuttgart, and Ludwig
Prandtl, professor of aerodynamics at the University of Göttingen, an
international
authority and pioneer in the theory of fl ight. Prandtl confessed that
he could not qualify as a practical expert because he had never actually
fl own,
but he could not resist to add, “When Professor Einstein, who certainly
has a
weaker knowledge of fl ight than I do, appears as an expert, that will
be very
interesting indeed.”?? In the end, the court picked a third person, Hans
Wolff ,
an engineer at the German Test Institution for Aviation (Deutsche
Versuchsanstalt
für Luft fahrt) in Berlin- Adlershof, as a fl ight expert partner of
Einstein.
In its session of January 7, 1922, the court solicited a comment on
Wolff ’s
opinion from Einstein, which he presented on January 18.?? Wolff ’s
opinion is
not available.
Einstein essentially maintained his earlier opinion that Drexler’s
invention
falls within the area of protection of Anschütz’s patent, even though a
British
patent from 1916,?? not known to him when preparing his fi rst opinion,
might
restrict this area. He proposed a formulation of the specifi c novelty
of
Anschütz’s patent that would stand even in this case as “a clearly
arranged
combination of a gyro pendulum and an apparatus for indication of the
direction
of the apparent gravity.”
The case ended with a settlement out of court, and Kreiselbau retracted
the
appeal