There have been five different types of recall advocated or practiced in America. They have been the recall of elected officers. The advisory recall of federal elected officers, the recall of judges, the recall of judicial decision involving the constitutionality of laws, and the recall of administrative officials. All but the recall of elected officers have been used sparingly and criticized severely. The recall of particular officers to stand for another election to the office which they held from a general election, began its modern phase with its adoption in the Los Angeles City Charter of 1903. Oregon in 1908 was the first to allow the recall of any elective public officer of the state. By 1920, the recall movement had reached its peak of legal acceptance and party and popular approval. Though it is still the law in many jurisdictions, it characterized the reforming zeal of fifty years ago rather than of the present.
The recall, like virtue, has been more preached than practiced. Assuming that the instruction of representatives was more democratic and right than the independency of representatives, the proponents of the recall, finding that the curve of independency during occupancy of office reached its peak in the middle of the term, thought that the recall might break up the curve. Woodrow Wilson called it: "the gun behind the door," but in practice the "gun" was cumbersome, complicated and required many people to fire it. Questions of spending, honesty and other simple moral issues achieved considerable response from the public. In 1942 twenty-seven states possessed general laws providing for the recall. Twelve of these allowed the recall of elected state officers. Additional states, not having general laws, allowed city charters to include recall provisions of passed special laws for the recall. Though most of this came before 1925, in the first quarter of the century approximately one hundred and fifty instance of the recall occurred. Only two of these had to with officers elected in state-wide elections. Those threatened by a recall election usually triumphed.
The recall therefore, though in principle a pure manifestation of the psychology of direct democracy, can hardly be said to have restored the direct communication between elected and electors so dear to the heart of direct democrats. It is possible, if the complaints of many politicians are even partially believed, that the recall was another step toward preventing the representative from acquiring an integrated morale in the performance of his specialized tasks. Again we have a case where increasingly complex politics is faced with a simple device aimed at complete control.
The initiative and referendum, like the recall, have been hinted at throughout the history of direct democracy's attempt to dominate representative democracy. But again like the recall, their period of greatest public discussion and enactment came in the late nineteenth century in America, carrying over into the first part of the twentieth century. Extreme proposals are associated with extreme conditions, and the initiative and referendum were a movement to bring government back under the control of the "incorruptible" people, they were part of the general protest against legislative corruption and the over-representation of large-scale industry. All the previous representative devices which might have secured a government in the popular image were to be pushed further toward the logical goal of direct action by the community on all matters affecting the community.
A survey by Stuart A. Rice of the number of articles on the initiative and referendum listed in the Reader's Guide for each year from 1900 to 1922 showed an increasing development of attention to the subject from 1900 to the years 1911 and 1912. In the early 1920's, attention dropped to an average below that of 1900. The number of states adopting measures providing for the two devices followed the same pattern, reaching a peak in 1912. The first success of the movement came in South Dakota in 1898 and forty years later some twenty states possessed the initiative and referendum. Some form of direct participation in lawmaking is found in every state save Delaware and there have been few extensions of the idea since World War I, although hundreds of elections involved its use. South Carolina in 1924 voted on fifty-two measures at one time. In California, where 173 measures were submitted to the electorate in twelve years, the highest voter-participation was found on measures relating to Bible reading, the sale of oleomargarine, the gasoline tax, the repeal of prohibition, the repeal of a law legalizing boxing, prevention of cruelty to animals, daylight saving, Sunday-closing laws, and unemployment relief. The least interest was shown in measures relating to private corporations, methods of voting in local referendums, a judicial council, election of judges and a governor's council, municipal and county charter amendments, and intergovernmental arrangements. Thus technical matters of considerable importance are slighted while morals are emphasized. Questions of public morals showed a participation of 82  per cent, question of public debt 45.7 per cent. On the average participation was less than 50 per cent.
The experience of particular states, California and Michigan, corresponds to a general observation in all states. V.O. Key and W.W.Crouch, writing about the California experience, remarked:
"...that initiators of proposition have usually been pressure organizations representing interests - commercial, industrial, financial, religious, political -- which have been unable to persuade the legislature to follow a particular line of action."
This situation has led to attacks on the claim that popular voting on issues created a feeling of community solidarity. A. L Lowell suggested that, since a minority most frequently initiated legislation, questions of a delicate moral and sectarian nature might be raised. In such an event, social peace would hardly benefit. Again, since participation is so small, the desire for "unity" on the part of the whole people is questionable. And the split in the final voting on a measure is bound to evidence disunity rather than unity. Though a large number of people may feel more contented with their government if they overtly manifest their power over it, a minority might well feel safer in the hands of representatives who are always in need of the support of 2 or 3 per cent of the electorate.
A single issue may split a community more than a tally of averages on many issues. Whereas a single issue may isolate a single enemy and thus draw the lines for social conflict, many issues conjure up many enemies and, correspondingly, many friends.
Since the initiative and referendum were proposed initially to push through popular legislation against the vested interests and their "controlled" legislators, it might be expected that large majorities would have characterized voting on initiated or referred measures. This did not occur. The most important measures were those to which most people were indifferent and regarding which the legislature was not under fire for inactivity. Hardly any striking examples of legislative conservatism have been expunged by the direct legislation of the people. Professor Aikin, after his study of the Californian experience, reduced the defense of the initiative to a psychological sentiment of power. Californians resented defective legislative apportionment less than the people of Illinois because the initiative might have been used to change the apportionment in California, but not in Illinois.
If we accept his statement of the difference in attitude and the explanation of it, it would seem then that the initiative and referendum foster a frame of mind that corresponds badly to the realities of political life. If some rusted implements are kept in a firehouse and the belief in them is so great that new implements are rejected, the firemen are not likely to quench an outbreak of fire.
Although the initiative, referendum and recall have not had remarkable success, the ideal of the instructed representative is one of the principal components in the mind of many Americans. A Fortune survey of 1938 asked a representative group: "Do you believe that a Congressman should vote on any question as the majority of his constituents desire, or vote according to his own judgment?" Thirty-seven and four-tenths percent indicated that he ought to follow the desire of the constituency. 54.1 per cent that he should use his own judgment and 8.5 per cent had no answer. Another study, by L. E. Gleeck, of the attitudes of Congressmen toward the opinions of their constituents declared that "probably more than half of the Congressmen voted against the expressed opinion of their constituents as judged by the mailbags." The same study, interviewing thirteen Congressmen on their views, found that twelve out of the thirteen asserted that their own judgment was the basis for their voting. At the same time, they stated that the opinions of their constituents were a heavy secondary influence, along with the views of other public leaders, the position of their political party, and public-opinion polls. Studies by F. V Cantwell and others of the effects of direct communications to representatives from the constituents by mail, telegraph, or petition point to no more than a professed desire to "take into consideration" all of the elements pertinent to the communications in forming a judgment on an issue.
The development of public-opinion polls, however, which grew up in the decades between the two World Wars, had some meaning to those who continued to see in direct, mass representation the solution of the problem of representation. When it became apparent that the polls which were being conducted really defined the momentary opinion of the electorate on a specific issue, the champions of popular representation took heart. Articles appeared proclaiming the coming of a new age of democracy. No longer would the individuals composing society be hampered in the free expression of their ideas. Conformance of the representative to the desires of his constituents would now be inevitable. In its first stage, the dispute over public opinion polls concerned their possible error. Many people who did not believe in instructions challenged the ability of the polls to reflect opinion exactly. But improved methods of polling soon destroyed that objection. A scientifically conducted poll, free from corruption, can more accurately depict the electorate's simple preferences on an issue than can an initiated or referred law, because the latter is always influenced by machine politics and extent of participation.
But the final meaning of the controversy over public-opinion polls is essentially a struggle between ideologies of representation. The basic criticisms of these polls are the same as the criticisms of the free operations of the majority of individuals on the government. The polls at best render only a limited, temporary, and forced verdict of the electorate. The straw poll exists in a political vacuum, like the utopias of history, unrestrained and undirected by the urgent requirements of conference, cooperation, and the compromise characterizing the legislative process. It is possible that out of the techniques of public-opinion polling a method of instructing representatives could be placed in operation. If such a procedure were legalized, direct representation of the traditional kind would achieve a signal triumph, but a revolution in the established representative system would result. At present, since the technicalities of polling are not popularly understood and the agitation for direct representation is not so great as it has been at other times, the poll has temporarily assumed an informational character. it is instructive to representatives who consider the opinions of constituents as one of a number of factors in making political decision.