On a few occasions, we have considered things like Truman being
assassinated in early 1948, or LBJ as well as JFK being fatally shot in
Dallas.  In such events, of course, the Speaker of the House--Joe Martin
in 1948, John McCormack in 1963--becomes President under the Presidential
Succession Act of 1947.

Or does he?  

What if the act is unconstitutional?  I first came across some plausible
arguments that it is in John D. Feerick, *From Failing Hands:  The Story
of Presidential Succession* (New York: Fordham University Press 1965),
pp.267-8, who in turn relies on Ruth C. Silva, "The Presidential
Succession Act of 1949," 49 Michigan Law Review 451 (1949).  While
googling references to Feerick and Silva, I came upon a detailed statement
of the case against the Act by Akhil Reed Amar and Vikram David Amar, "Is
the Presidential Succession Law Constitutional?", 48 Stanford Law Review
113 (1995):

http://islandia.law.yale.edu/amar/lawreview/1995Succession.pdf

Essentially, there are at least three arguments that the 1947 Act is
unconstitutional:  

(1) Are the Speaker and the President pro tempore "Officers" in the
constitutional sense?   Article II of the Constitution states:

"In Case of the Removal of the President from Office, or of his Death,
Resignation, or Inability to discharge the Powers and Duties of the said
Office, the Same shall devolve on the Vice President, and the Congress may
by Law provide for the Case of Removal, Death, Resignation or Inability,
both of the President and Vice President, declaring what Officer shall
then act as President, and such Officer shall act accordingly, until the
Disability be removed, or a President shall be elected."

If "officer" here is shorthand for "officer of the United States"--and the
Amars present strong reasons to think it is--it is very doubtful that the
Speaker and President pro tempore are "officers" in that sense (though
they are of course officers of the House and Senate, respectively).  
Indeed, the "Incompatibility Clause" of Article I, Section 6 makes clear
that sitting members of Congress cannot hold "any Office under the United
States."

(Of course if one wants to interpret the word "officers" broadly enough,
it could include not only officers of the US but officers of the Congress
and indeed *state* officers!  After all, as the Amars note, "Article VI's
reference to 'Officers' ranges beyond federal officials, and explicitly
includes 'Officers...of the several States.'"  ObWI: Congress decides that
the governor of whatever is the most populous state at the time is next in
line after the VP...)

You may say that this argument also holds against the 1792 Succession Act,
which puts the President pro tempore of the Senate next in line after the
Vice-President. (The 1886 Act is free from any such problem--it goes from
Vice-President to Secretary of State, and does not include any member of
Congress in the line of succession.) Indeed, it does--and James Madison in
a letter to Governor Edmund Pendleton of Virginia, objected to the Act for
precisely this reason:  

"On another point the Bill certainly errs. It provides that in case of a
double vacancy, the Executive powers shall devolve on the Presidt. pro.
tem. of the Senate & he failing, on the Speaker of the House of Reps. The
objections to this arrangement are various. 1. It may be questioned
whether these are officers, in the constitutional sense. 2. If officers
whether both could be introduced. 3. As they are created by the
Constitution, they would probably have been there designated if
contemplated for such a service, instead of being left to Legislative
selection. 4. Either they will retain their legislative stations, and
their incompatible functions will be blended; or the incompatibility will
supersede those stations, & then those being the substratum of the
adventitious functions, these must fail also. The Constitution says,
Congs. may declare what officers &c. which seems to make it not an
appointment or a translation; but an annexation of one office or trust to
another office. The House of Reps. proposed to substitute the Secretary of
State, but the Senate disagreed, & there being much delicacy in the matter
it was not pressed by the former."

http://press-pubs.uchicago.edu/founders/documents/a2_1_6s3.html

(2) Does the Speaker (or President pro tem) have to resign his
Congressional position upon his succession to the presidency?  The 1792
Act was silent on this.  The 1947 Act is unambiguous: "the
Speaker...shall, upon his resignation as Speaker and as Representative in
Congress, act as President."  There is a similar provision for the
President pro tempore.  This runs into Madison's objection, stated above:  
the language of the Constitution seems to require that the powers and the
duties of the presidency be *annexed* to another office. This was Senator
Hoar's position during the debate on the 1886 Act:

"[T]he Presidency is annexed by law to an office.  It is not a person
holding an office at the time succeeding to the Presidency, but it is an
officer continuing in that office who is to perform as an annex or
incident merely to another office the great duties of the Presidency
itself."  In other words, if the Speaker or President pro tem resigned, he
would have no office to which the function of acting as President could be
attached.  But if he *didn't* resign that would be a clear violation of
the separation of powers--to have a member of Congress acting as POTUS.  
Moreover, his tenure as acting President could be ended by expulsion from
whichever House he was a member of, or by loss of his legislative seat at
the polls.  (By contrast, Madison saw no conflict in the Secretary of
State retaining his position and acting as President simultaneously.)  

As Feerick notes (p. 266):  "In rebuttal, it has been contended that the
Constitution merely requires the successor to be an officer of the United
States at the time he begins to act as President.  Hence,if the officer
resigns at the time he takes the presidential oath, this requirement is
satisfied.  The flaw in this position is that the Constitution
specifically provides that the officer shall act as President 'until the
Disability be removed, or a President shall be elected,' implying that he
is to remain an officer..."

At first sight, this may seem analogous to a similar dispute about the
Vice-President becoming President.  John Quincy Adams and others insisted
in 1841 that Tyler was properly styled "Vice-President, acting as
President."  Tyler, of course, insisted that the Vice-Presidency had
become vacant and that he was now President, period.  (The White House
returned all mail addressed to "Acting President John Tyler" on the ground
that there was no such office.)  And this precedent of course prevailed.  
But as Tyler's defenders noted at the time, there is a difference between
the constitutional language for succession by a Vice-President and by an
"officer."  In the latter case, the Constitution is explicit that such a
person merely *acts* as President.  By contrast, it states that "In Case
of the Removal of the President from Office, or of his Death, Resignation,
or Inability to discharge the Powers and Duties of the said Office, the
Same shall devolve on the Vice President" and this is at least open to
Tyler's interpretation that "the Same" refers to the "Office" not merely
its "Powers and Duties."  In any event, this dispute was finally resolved
by the Twenty Fifth Amendment, which says flatly that in the case of the
President's death or removal from office, "the Vice President shall become
President.")

(3) A third legal objection to the Act of 1947 is its so-called bumping
provision.  That is, if a cabinet officer becomes President (because there
is no Vice-President, Speaker or President pro tem at the time) he acts as
President until a President, Vice-President, Speaker or President pro tem
is available.  Thus, a Secretary of the Treasury who acts as President
cannot be superseded by a Secretary of State but he can be by either a
Speaker or a President pro tempore.  This seems to contradict the
Constitution's provision that the officer appointed by Congress shall act
"until the Disability be removed, or a President shall be elected."  
(Incidentally, the Act makes it clear that only a person constitutionally
eligible to be Presdident can act as President.  It would be interesting
to formulate a scenario where Henry Kissinger is next in line--and
challenges that provision of the Act in *Kissinger v. Shultz* or
*Kissinger v. Simon* arguing that the Constitution says nothing about the
officer *acting* as president having to be a natural-born citizen...)

Apart from narrow objections based on the text of the Article II
provision, there are also broader "strucural" objections to having members
of Congress in the line of succession, relating for example to the danger
of conflict-of-interest in presidential and vice-presidential impeachment
proceedings.  This point, much emphaszed by the Amars in their article, is
hardly an academic one, as next-in-line Benjamin Wade's participation in
the Andrew Johnson impeachment trial shows.  Even Wade's recusing himself,
though, would not have solved the problem--because quite apart from the
ethical problems of Wade's own self-interest, putting a Congressional
leader in the line of succession puts the US closer to a "parliamentary"
model (which the framers of the Constitution consciously rejected) under
which Congress can choose its own leader to head the government.  In other
words, there was a conflict of interest not just for Wade, but for
Congress as a whole.

OK, so there are the objections. (BTW, I don't want to consider the impact
of the 25th Amendment on any of the questions here, so I am dealing solely
with the period before it was ratified.)  Note that most of these problems
would  not exist with a Cabinet line of succession (as under the 1886
Act):  Cabinet officers are unquestionably officers of the United States,
and if they retained their offices while acting as President there would
be no separation-of-powers question.  A deputy could supervise the work of
the department whose head might be serving as President.

But to get back to the 1947 Act--the problem is that while there are good
arguments it (like that of 1792 but *not* that of 1886) is
unconstitutional, how in practice could this be established?  Who would
have standing to sue (acting) Presidents Joe Martin or John McCormack and
tell them that they were *not* validly in the line of succession?  
Presumably Secretary of State George Marshall in 1948 or Dean Rusk in 1963
would be the logical plaintiffs, claiming that they were next in line.  
But it is of course extremely difficult to see them doing so, and if they
did, the Supreme Court would probably find a way to dodge the issue--say,
by declaring it a "political question."   I mean, after all, who can
imagine the Supreme Court deciding who is to be President?  :)  

(One other what-if here:  According to Feerick, p. 208, Congress *almost*
made the President pro tempore, not the Speaker, next in line in 1947.
Richard Russell introduced an amendment to that effect.  It was defeated
55-31 largely because Vandenberg, who was then President pro tempore,
opposed it, arguing that the Speaker was "the officer reflecting the
largest measure of popular and representative expression at the instant
moment of his succession."  Suppose Russell's amendment had passed.  The
US would be a big step closer to having Walter George, Richard Russell,
Allen Ellender, Strom Thurmond--perhaps as a very old man but perhaps
instead as the spry 78 year old of 1981--John Stennis, James Eastland,
Robert Byrd, or the very elderly Carl Hayden as acting POTUS.  Or would
the Senate change the way it elected a President pro tem--deciding not to
rely entirely on seniority any more--if that office were next in line of
succession after the VP?)

--
David Tenner
dten...@ameritech.net