When do Presidents Choose "Extreme" Nominees?

One of the questions being pondered this summer by the left side of the legal blogosphere is whether President Obama was wise to choose as his first Supreme Court nominee the apparently more cautious Sonia Sotomayor rather than a "liberal lion" in the mode of William Brennan.  This discussion led me to ask a question of my own: when is a President likely to select an "extreme" nominee rather than a more moderate one?

The below chart uses the Segal/Cover Ideology Scores to explore this question. Jeff Segal and Albert Cover devised these scores by using newspaper reports to measure the perceived ideology of a nominee at the time of his or her nomination. They provide a score for every nominee named between 1937 and 2006.

If we consider an "extremist" to be any nominee falling more than one standard deviation outside of the median, then nominees listed above the first dividing line in the chart are the conservative "extremists", while those below the second line are the liberal "extremists" (remember, the Segal/Cover scores measure perceived ideology at the time of the nomination - which perhaps accounts for Justice Blackmun's score).

The chart does not yield an easy answer to my question, but it does show that at least one frequently told story about extreme nominees is not true: a friendly Senate is no guarantee that a president will choose an extreme nominee. Of the 22 extreme nominations, eleven were made during unified government and eleven during divided government.

	Score	President	Pres	Senate	Vote
D. Ginsburg	0	Reagan	R	D	n/a
Scalia	0	Reagan	R	R	98-0
Carswell	0.04	Nixon	R	D	45-51
Rehnquist, AJ	0.045	Nixon	R	D	68-26
Rehnquist, CJ	0.045	Reagan	R	R	65-33
Bork	0.095	Reagan	R	D	42-58
Alito	0.1	GW Bush	R	R	58-42
Burger	0.115	Nixon	R	D	74-3
Blackmun	0.115	Nixon	R	D	94-0
Roberts, CJ	0.12	GW Bush	R	R	78-22
Haynsworth	0.16	Nixon	R	D	45-55
Thomas	0.16	GHW Bush	R	D	52-48
Powell	0.165	Nixon	R	D	89-1
---------------------	-----------	-------------	-------	-----------	----------
Stevens	0.25	Ford	R	D	98-0
Miers	0.27	GW Bush	R	R	n/a
Burton	0.28	Truman	D	D	Voice
Stone, CJ	0.3	FDR	D	D	Voice
Souter	0.325	GHW Bush	R	D	90-9
Byrnes	0.33	FDR	D	D	Voice
Kennedy	0.365	Reagan	R	D	97-0
O'Connor	0.415	Reagan	R	R	99-0
Breyer	0.475	Clinton	D	D	87-9
Clark	0.5	Truman	D	R	73-8
White	0.5	Kennedy	D	D	Voice
Whittaker	0.5	Eisenhower	R	D	Voice
Frankfurter	0.665	FDR	D	D	Voice
RB Ginsburg	0.68	Clinton	D	D	96-3
Minton	0.72	Truman	D	R	48-46
Reed	0.725	FDR	D	D	Voice
Douglas	0.73	FDR	D	D	Voice
Vinson, CJ	0.75	Truman	D	D	Voice
Warren, CJ	0.75	Eisenhower	R	R	Voice
Stewart	0.75	Eisenhower	R	D	70-17
Goldberg	0.75	Kennedy	D	D	Voice
---------------------	-----------	-------------	-------	-----------	----------
Fortas, CJ	0.845	Johnson	D	D	n/a
Black	0.875	FDR	D	D	63-16
Harlan, John M	0.875	Eisenhower	R	D	71-11
Murphy	1	FDR	D	D	Voice
Jackson	1	FDR	D	D	Voice
Rutledge	1	FDR	D	D	Voice
Brennan	1	Eisenhower	R	D	Voice
Marshall	1	Johnson	D	D	69-11
Fortas	1	Johnson	D	D	Voice

Comments

[Rob Robinson]

Might this be primarily a function of what, exactly, a president wants in a nominee? If the nominee is chosen to help fulfill a campaign promise (O'Connor) or to court an influential voting bloc (Sotomayor), advancing ideology may be a secondary goal. Choosing a moderate may thus be preferable because it helps ensure the primary goal is met.

It also might be interesting to see the president's approval ratings at the time of the nomination, since it might be harder to push through a more ideological nominee when the president is weaker (i.e. Souter). I suspect it might lead to the same sort of non-pattern that we see above, but still worth a look.

[Lori Ringhand]

By my quick count, there were 43 total nominations in this time period: 18 during divided government and 25 during unified government.

[Alan E. Dunne]

But how many total nominations were there in divided v. unified(partywise) President/Senate pairs?