Richard Morris Hunt, William K. and Alva Vanderbilt House, New York, NW corner of Fifth Avenue and 52nd, 1879. (Online)
It is now time to go back and look at what happened in New York prior to January 1891. I will specifically discuss the resurrection of the “Anciens” in the work of McKim, Mead, & White later in this chapter, but first, I must review what the two leading Eastern architects, Hunt, and his former pupil, Post had been designing at the same time that Root was designing the Rookery and the Monadnock. Hunt, educated at the École des Beaux-arts (1843-54) during the Neo-Gréc rebellion of Henri Labrouste, had never veered far from his roots. The Francois I or French Renaissance style had been given its initial stamp of approval in 1879 by New York’s trend-setter, Alva Erskine Smith Vanderbilt (also French-educated) with her (Mrs. William K.) $3 million house designed by Hunt that opened to New York High Society on March 26, 1883. In the 1885 A.I.A. survey of American architects’ favorite buildings, Hunt’s Vanderbilt house was rated third (following Richardson’s Trinity Church and the U.S. Capitol). Alva Vanderbilt’s mansion had set the mark for stylishness in New York during the 1880s, and that would be Francois I.
Richard Morris Hunt, William Borden mansion, Chicago, 1884. (Stein, Hunt)
Hunt (who had designed Marshall Field’s house on Prairie Avenue ten years earlier) was commissioned a year later by William Borden (the son of John Borden, owner of the Borden Block designed by Adler, that begs the question why did he not hire Adler to design the house?) to design a similarly styled Francois I mansion at the northwest corner of Lake Shore Drive and Bellevue Place (the southern-most lot of Potter Palmer’s new residential neighborhood along Lake Shore Drive: Palmer lived six blocks to the north) that was completed in 1889. By this date, however, Hunt, working with Frederick Law Olmsted, was already deep into the Francois I design of what was to be an even more extravagant estate than that of Alva and William K. Vanderbilt: the Biltmore in Asheville, N.C, for William’s youngest brother, George Washington Vanderbilt.
Left: Richard Morris Hunt, Biltmore, Asheville, NC, 1889; Right: John Singer Sargent, George Washington Vanderbilt, 1890. (Online)
1.4. THE DAKOTA
Henry Janeway Hardenbergh, The Dakota Apartments, New York, Central Park West and 72nd, 1882. (Stern, New York-1880)
Such was also the case with the Dakota Apartment House built in 1882 by Edward Clark, the president of the Singer Sewing Machine Company, as part of his speculative strategy to promote the upper West Side as an upper middle class residential neighborhood. Clark’s vision for the new project was: “There are but few persons who are princely enough to wish to occupy an entire palace, but there are many who would like to occupy a portion of a great building, which would be more perfect in its arrangement than any palace in Europe.” Clark was taking a double gamble with the Dakota. First, there was no guarantee that this location would prove to be desirable in the eyes of his target market, and second, he was introducing New Yorkers to a different type of housing unit. Instead of the fashionable row or townhouse, Clark thought that New Yorkers would take to the Parisian courtyard apartment house.
Henry Janeway Hardenbergh, The Van Corlear, New York, 7th Avenue and 55th Street, 1878, (Online)
Clark hired architect Henry Janeway Hardenbergh to design New York’s first upscale, multistoried apartment house based on this model, for a block on the west side of Central Park West between 72nd and 73rd, that would have an unobstructed view of Central Park. Hardenbergh, who had apprenticed under Detlef Leinau (v.3, sec. 5.3), had already designed an earlier project for Clark, the Van Corlear that had quickly gained the reputation as being the first New York apartment house that felt and functioned like a true Parisian courtyarded apartment house. The Van Corlear, however, did not yet “look the part.” Hardenbergh succeeded in making the Dakota “one of the most perfect apartment houses in the world,” as it was judged upon its completion in September 1884. No expense was spared in its design and construction; it was as good as the quote above made it out to be.
Hardenbergh, The Dakota Apartments. Birdseye view of court. (Online)
It is obviously the Dakota’s two-story Francois I châteauesque roof that I am interested in. On the Central Park West elevation, he topped the corner pavilions with sharply profiled gables that were linked to the central pavilion’s gable by a transverse mansard roof. He then activated the roof’s silhouette with the requisite pinnacles, dormers, and chimneys, lending an air of true French sophistication to its eight-story visage as it raised its head above the tree line of Central Park. Its private dining room at the time was known to be “the handsomest dining room in Manhattan.” If this wasn’t sufficient to pique Root’s interest during his business trips to New York during the consolidation process, the building’s clientele, known to be among Manhattan’s culturally elite, especially in the music world, would have, more than likely, offered Root the opportunity to visit the building on more than one occasion.
Hardenbergh, The Dakota Apartments. View from Central Park. (Stern, New York-1880)
FURTHER READING:
Alpern, Andrew. The Dakota. New York: Princeton Architectural Press, 2015.
Stern, Robert A.M. New York: 1880. New York: Monacelli Press, 1999.
Burnham & Root, Left: W.C.T.U. Woman’s Temple, Chicago, SW corner of La Salle and Monroe, 1890. (Hoffmann, Meanings); Right: Masonic Temple, Chicago, NE corner of State and Randolph, 1890. (Hoffmann, Root)
In addition to the Chicago Hotel, Burnham & Root had two other mammoth projects to design in early 1890 whose owners wanted to cash in on the free publicity and exposure of their cause that the World’s Fair could generate. In addition to having to completely redesign the W.C.T.U.’s Woman’s Temple so that it could be built on the site of Field’s stillborn office tower, Chicago’s Freemasons had also hired the firm to design a new headquarters/temple on the northeast corner of State and Randolph. As both projects were being designed by Root at the same time, and that they were both not merely speculative office buildings, but headquarters for major social organizations, we can compare the design of these two skyscrapers to discern where Root’s design ideas had evolved to as he moved into the new decade.
The first thing that one is immediately struck by is the addition of a roof to Root’s new skyscrapers. All skyscrapers designed by Root prior to 1890, with the sole exception of the tower in the Kansas City Board of Trade and the pinnacles atop the Rialto’s piers (unless we count his entry for the six-story Cincinnati Chamber of Commerce), no matter whether the building’s function was a hotel, a company headquarters, or a speculative office building, were flat-topped, urban palazzos. (And once again, I repeat my earlier assertion that Root, and not Richardson, had brought the box to Chicago.) The Chicago Tribune had noticed this sea-change immediately in praising the Woman’s Temple’s design as being close to “the ideal of office buildings. It appeals to the eye in a manner wholly unlike any of the ‘dry-gods box’ giants which are becoming too common.” How are we to account for this abrupt change in Root’s aesthetics as we find a roof added not only to these first two buildings of 1890, but also to his design for the San Francisco Examiner and an unidentified skyscraper being designed as the first two buildings were being finalized?
Burnham & Root. Left: San Francisco Examiner Project, San Francisco, 1890. (Monroe, Root); Right: Project for a Skyscraper, 1890. (Hoffmann, Root)
CHAPTER 1: EVENTS IN NEW YORK
1.1. THE CONSOLIDATION OF THE A.I.A. AND W.A.A. BRING BURNHAM AND ROOT TO NYC
In November 1889, Burnham and Root had successfully completed their campaign to reform the old A.I.A. While Root had represented the W.A.A. in these negotiations, the two partners had agreed that Burnham, following his defeat in the presidential election at the 1888 W.A.A convention, would play the Trojan Horse from within the A.I.A. and therefore, had distanced himself from the W.A.A., in order to appear to be more committed to A.I.A. issues. On November 20-1, 1889, both the W.A.A. and the A.I.A. had met in a joint convention in Cincinnati to approve the final consolidation of the two regional professional organizations into a new, nation-wide Institute. While it was finally argued successfully that legally the name of “The American Institute of Architects” had to be maintained in order to avoid all of the legal headaches in applying for a new state charter, there was no doubt that the West had won the war in that it was the W.A.A.’s constitution and professional philosophy that were the foundation for the new Institute. While Richard Morris Hunt was respectfully recognized as the Dean of American architects by being elected the first President of the new Institute, Root’s (note it was not Burnham) central role in the overall merger was rewarded with his election as its Secretary, the second highest position in the new organization. All official correspondence of the new A.I.A., therefore, would no longer go through New York, but through the top floor of the Rookery in Chicago. Root’s election as Secretary was also tantamount to the certainty that following Hunt’s two-year term, Root would be nominated to fill the presidency at the 1891 convention. (Root had, thus, been “knighted” as the Number Two Architect in the U.S.)
1.2. NEW YORK CITY PREPARES FOR THE 1892 FAIR
While both Root and Burnham had played major roles in the consolidation negotiations, spending many days in New York hammering out the final details with their opposites, New York was at least as busy as Chicago in trying to convince Congress to make it the host city for the 1892 World’s Fair. In fact, Burnham and Root, who were personally involved with both of these tightly contested campaigns, that is, the reform of the A.I.A. and the contest to have their city awarded the Fair that were taking place simultaneously, (the decision naming Chicago was made on February 24, 1890, exactly three months after the formation of the new A.I.A. was approved in Cincinnati) would spend many days in New York during the last quarter of 1889 (providing them with a legitimate cover for spying on New York’s plans for the Fair). It is now time to examine what new buildings they saw in New York and how they may have been influenced.
Chicago may have had a head start in “gussying itself up” to convince the country’s political leaders to name it as the host of the Fair simply because it had been playing catch-up in construction ever since the 1871 fire, (i.e., the Interstate Exposition Building, the Board of Trade-then the tallest building in the U.S.- and of course, the Auditorium), once the idea of an American World’s Fair gained traction, New York shrugged off its smugness, took an honest look at itself, and dove into the competition with as much “Old and New Money” as was needed to erect a number of new buildings to respond to the “Cowtown in the West.”
George B. Post. Left: Mills Building, New York, 1881; Right: Produce Exchange, New York, 1880. (Online)
The last important building in New York I have reviewed was George Post’s Produce Exchange of 1880-3. As construction in Chicago had boomed with the economy in 1881-1885, a similar phenomenon had occurred in New York. While Richardson had always given his buildings a picturesque roofline until 1885 with his first box, the Field Wholesale Store, the less expensive flat-roofed box topped with a straight-lined cornice had been in fashion for tall buildings, since the start of the depression of 1874. This had continued into the first half of the 1880s, as best represented by Post’s Mills Building and Produce Exchange in New York, and Root’s Burlington and Rookery Buildings in Chicago. As the national economy continued to improve, however, it was inevitable that as the economy improved, more ornate silhouettes would become more affordable and once again be demanded by clients. In v.5, sec. 1.12, I stated that the American version of laquerelle des Anciens et des Modernes in architecture had sprung to life around 1886, first with the death of Richardson on April 27, 1886, and second, with the corresponding design in 1887 of McKim, Mead, & White’s Boston Public Library. This is how I described this divergence between the East and the West in the mid-1880s:
“Precisely at this moment, however, Eastern American architects had also consciously diverged from the American omnipresent Romanesque Revival (Richardson had died on April 27, 1886, only a week before the Haymarket Square bombing). However, rather than looking to the problem for the solution, these architects chose to repurpose (positive)/ imitate (negative) the Classical architecture from the past. The best example of this sea change in Eastern architectural style was the new Boston Public Library designed in 1887 by McKim, Mead, & White, that poignantly sat facing the great Richardson’s Trinity Church. I will (need to) discuss the reasons for this change of style in depth prior to discussing the 1893 Columbian World’s Exposition. For now, it must suffice to state that as Chicago’s architects, led by Burnham and Root, had diverged from their East Coast contemporaries in terms of professional practice with the formation of the W.A.A. in November 1884, it should not surprise us that they would also take their own architectural path into the future. These divergent paths will collide in January 1891…”
Burnham & Root, Chicago Hotel, northeast corner of Dearborn and Jackson, 1889. The 1895 office/theater addition is at the rear. (Online)
In early 1890 Chicago was engaged in a no-holds–barred, winner-takes-all battle with New York City to host the 1892 World’s Fair to commemorate the 400th anniversary of Columbus’ discovery of the New World (see Vol. Seven). If Chicago succeeded in winning the contest, the city would need a series of large hotels to provide rooms for the Fair’s visitors. A group of investors who intimately understood the potential profit from such an investment were planning to build a hotel on the northeast corner of Dearborn and Jackson, that only a year earlier had been slated for the W.C.T.U.’s new headquarters (see v. 4, sec. 5.4). The site was owned by Eugene S. Pike, who once the W.C.T.U. had moved its planned building to Marshall Field’s vacant lot on La Salle, had contemplated building a 16-story office building designed by Burnham & Root. That is, until Chicago’s prospects for winning the Fair had markedly improved in late 1889. Pike had insider knowledge of the City’s Fair proposal because he was a member of the City’s Committee of 100 appointed by then Mayor DeWitt Creiger in July 1889 to promote Chicago’s Fair submission. In addition, Burnham and Root had already been “unofficially” working with a number of the Committee “behind the scenes” to assist the effort before the city even formally announced that it was in the running.
In fact, Burnham organized this group of investors that in addition to Pike and Root, also included George Fuller and Norman Ream. This was basically the same group, minus Peter Brooks, who had financed and constructed the Rookery. These investors were savvy enough to know that in order to be open and ready for the start of the Fair, the plans for the new Chicago Hotel would have to be completed and be ready to go out to bid on the first day after the final decision had been made. Therefore, Burnham saw to the incorporation of the company in November 1889, some three months before Chicago was formally named by Congress on February 24, 1890. It may have just been a coincidence that the hotel was incorporated only a few weeks before the Opening Night gala of the Auditorium’s theater on December 9, but I have to speculate that Burnham and Root wanted a modicum of revenge for having lost this commission directly through the intervention of Ferdinand Peck. The Auditorium’s hotel was scheduled to open the following month in January. It was built with ten floors; the Chicago Hotel was to have 14 floors. But what really had made the Auditorium Hotel an anachronism before it even opened was that American tastes in hotel rooms had changed between 1887 and 1889. The Auditorium Hotel was designed on the European model of shared/public bathrooms, typically one per every ten hotel rooms (just down the corridor). If you look carefully at the floor plan of the Chicago Hotel, every hotel room has been designed with its own bathroom. The Auditorium Hotel would never receive the prestige Peck had hoped for. (Burnham completed his “Auditorium alternative” in 1895 with an addition that included a 2000-seat theater and a 16-floor office annex to the east of the hotel. Its name was correspondingly changed to the Great Northern Hotel and Theater.)
Burnham & Root, Chicago Hotel. Typical floor plan. Note that the rooms show a separate bathroom, (Hoffmann, Root)
Ream was one of the owners who had erected the Midland Hotel in Kansas City in 1886, also designed by Burnham & Root. In fact, one can understand the Chicago Hotel as an extruded version of the Midland Hotel. Each hotel had a U-shaped plan, with double-loaded corridors on the three street fronts that wrapped around an exterior lightcourt. As a double-loaded corridor scheme, like the Midland Hotel, the Rookery, and the Rand-McNally Building, the skylight was brought down to the third floor. Under this Root located the hotel’s service counters and offices. Although Root had included in the light court, his, by now, signature cantilevered semi-circular stairway (used in the Rialto, the Insurance Exchange, the Rookery, and the Midland Hotel) the surviving photos of the atrium do not show the stairway intersecting with the skylight.
Burnham & Root. Left: Midland Hotel (Hoffmann, Root); Right: The Rookery.
Burnham & Root, Chicago Hotel. Above left: The skylight and exterior lightcourt, that once again, as Root had done with the Rookery (above right), was given an expression in which the horizontals, and not the verticals, were continuous. Here one can see the difference in the required daylight between a hotel room and an office. (Left: Hoffmann, Root); Below: The hotel’s atrium: note that it does not appear that the oriel window/stair in the lightcourt was brought down through the skylight as was done in the Rookery. (Left: Online; Right: Hoffmann, Root)
As the site was diagonally across the intersection of Dearborn and Jackson from the Monadnock, Root again had an opportunity to design an urban ensemble, similar to the portal to the Board of Trade he had designed with the Insurance Exchange and the Rookery at the intersection of La Salle and Adams. Indicative of the fact that Burnham and Root were intimately connected with both sides of the urban battle between La Salle and Dearborn, the portal created by the new Chicago Hotel and Monadnock Block framed the tower of the Dearborn Station.
Intersection of La Salle and Adams, looking south. The Rookery (left) and the Insurance Exchange (right) as the portal to the Board of Trade district. (Merwood-Salisbury, Chicago 1890)
Burnham & Root, Intersection of Dearborn and Jackson: Left-the cylindrical bay windows of the Chicago Hotel; Right-the Monadnock Block. The tower of the Dearborn Street Station is at the end of the vista. Note that the new U.S. Courthouse has completely filled the old Post Office Square. (Leslie, Chicago Skyscrapers)
Burnham & Root, Monadnock Block (left) and the Chicago Hotel (right) Note how the new Courthouse (Cobb, 1906) completely filled the old Post Office Square, eliminating the only public space in the South Loop. (Online)
This site also included the Post Office Square, the only remaining “public open space” beyond lake shore of Lake Park. Therefore, the open space on the west side of Dearborn would offer long vistas of both buildings, meaning they would be easily seen together, as an ensemble should.
Contemporary designs by Root at the southeast corner of the Post Office Square: Monadnock Block (#7) and the Chicago Hotel (#8). The Owings Building is #3, the Union League Club is #6. (Rand-McNally Views)
Burnham & Root had been designing the hotel roughly on a parallel timeframe with that of the Monadnock Block. What the Brookses had prohibited Root to do in the Monadnock he could experiment with in the design of the planned hotel, if for no other reason than he was one of the project’s owners. Root gave the hotel the same smooth masonry surface of the Monadnock so the two buildings would act as a group rather than two individual buildings.
Burnham & Root, Chicago Hotel. (urbanremainschicago.com)
Some historians have speculated that Root had used his first 16-story design of the Monadnock, the steel-framed version that Shepherd Brooks had rejected. I don’t think this was the case simply because this design was for an office building that required more window area than did a hotel. I see the Chicago Hotel’s elevation as the next iteration in Root’s exploration of the “continuous curtain wall.” In the Rand-McNally Building, Root had detailed its elevation as one continuous 10-story sheet of terra cotta. While the Monadnock was structured with masonry piers and not a steel frame, Root had still detailed its exterior as a continuous surface of brick. He treated the Chicago Hotel’s elevation as the third in this sequence of continuous brick surfaces. The Rand-McNally was a single plane, sandwiched by its neighboring buildings. In the Monadnock, he avoided sharp corners to achieve continuity by carving the corner with the ever-increasing chamfer. In the Chicago Hotel, as he had done four years earlier in the Pickwick Flats, he simply eliminated the sharp corner by placing a cylindrical bay window at the two corners. (Similar to how Starrett did in the Hyde Park Hotel.)
Above: Left: Rand-McNally Building; Right: Monadnock. Below: Left: Pickwick Flats, southeast corner of Michigan and Twentieth, 1886 (Hoffmann, Root); Right: Theodore Starrett, Hyde Park Hotel, Chicago, 1887. (Condit, Chicago School)
In the Chicago Hotel elevations Root, however, could not resist his propensity to achieve repose: a balance between the horizontal and the vertical. He articulated the general massing into a single-story base with a 12-story middle. He then stopped this continuous, 12-story tall surface at the 14th floor with a heavy, projected stringcourse that allowed the 14th floor to be read as a cornice. This consisted of a continuous line of repeated rectangular windows and small mullions that made the floor read as a void, and correspondingly allowed the parapet above the windows to appear to float, as it topped off the building. The parapet reprised the Monadnock’s coved cornice, with the only difference being Root set up the cute with a series of machiolations. This detail tied the two buildings together as an ensemble at their rooflines.
Left: Monadnock Block; Right: Chicago Hotel. (chuckmanchicagonosalgia.wordpress.com)
Unfortunately, Root still felt compelled to add a few, superfluous horizontal projections as a transition from the column-and-void base (à la Holabird & Roche) to the 12-story center. (This detail was similar to his technique of creating a transition story by taking the material of the upper floor but using the language of the lower one.) This reveals his hesitancy, even at this late date in his career, to make his skyscrapers “vertical:” we owe the verticality in the Monadnock to Peter Brooks’ command to eliminate all projections, not to Root’s final epiphany that a skyscraper could be a “proud and soaring” building.
Burnham & Root, Chicago Hotel. (urbanremainschicago.com)
Unlike the Monadnock, Root incorporated his old friend, the Romanesque semi-circular arch in the hotel’s elevations, but limited their use only to the twelfth floor, as a method of capping the building. The three street elevations alternated between bay windows (with single windows) and 11-story tall arched bays with inset paired windows and recessed spandrels, a detail taken from the Rand-McNally Building.
Burnham & Root. Left: Arched windows in the Rand-McNally Building; Right: Arched windows in the Chicago Hotel.
As was the case in the Rand-McNally Building, the arches’ location seem to be arbitrary. In addition to the segmented arches in these bays, he also continued the arch into the bay windows in the eleventh floor with a series of individual semicircular arches. One must wonder, however, why the arches in the bays were located in the twelfth floor rather than in the thirteenth, i.e., at their top? My guess is that Root wanted to make a transition between the fourteenth-floor cornice of flat-headed windows and the twelfth floor’s arches. He did this by detailing the thirteenth-floor with all flat-headed windows (from the floor above) but still kept the undulating surface of the floors below. While this “kinda works” above the arched paired windows, the semicircular arches in the curved bays needed to be relocated to the top floor (the 13th) of the bays: they just look out of-place where they ended up. This is quite evident because Root did exactly this (terminate the bays with the arches) in all of the polygonal bay windows. Again, one is tempted to assign these “design errors” to one of Root’s assistants following Root’s death in January 1891. However, this goes against how every draftsman in Burnham & Root described his design process of thinking for a few days, and then drawing the entire, completed design in one day,
Burnham & Root, Chicago Hotel. (urbanremainschicago.com)
Burnham & Root, Chicago Hotel. Typical floor plan. Note the two masonry party walls at the rear. (Hoffmann, Root)
While the building’s structure was completely iron-framed with a system of wind bracing that comprised of double diagonal tie rods that extended for two stories and passed through a column in the middle, it still relied on two masonry party walls at the rear, as required by the building code.
Burnham & Root, Masonic Temple. Cross section showing the configuration of the diagonal bracing that was similar to that used in the Chicago Hotel. (Engineering Record, January 21, 1893)
When we review the seven buildings erected in 1888-90 (Tacoma, Chamber of Commerce, Rand-McNally, Second Leiter, Manhattan, Monadnock, and the Chicago Hotel), only one of these, Jenney’s Manhattan Building appears to have been erected solely with a skeleton frame, but not because he wanted to be the first to do so. The other six were either required by the building code to have masonry party walls or the designers chose to incorporate lateral masonry walls to assist in stiffening the building against wind loads. I would credit this fact to:
-first: there appears to have been no rush, contrary to what historians like to record, to claim to have been the first architect to design a building erected solely with the iron skeleton. (Nowhere in any publication of the period does a writer claims that a building was the first to have been so constructed. It was, apparently, no big deal. Was this simply because Buffington’s patent was the first such building?) This, more than likely, was the result that iron framing was more expensive than a masonry wall, and no client in Chicago was interested in this extra cost;
LeRoy S. Buffington. Left: 28-story Cloudscraper, Minneapolis, 1888. (Inland Architect, July 1888); Right: Patent for Iron Building Construction, May 22, 1888. (Online)
-and second: Jenney was confronted in the design of the Manhattan with the fact that there were existing buildings on both interior sides of its lot, meaning that if he was to build the masonry party walls from the ground up, he would have to shore up the foundations of the adjoining buildings and reinforce their foundations. This was obviously more expensive in time and money than his final solution: to cantilever the structure’s beams at each floor to the lotline upon which he could then erect the requisite masonry party wall.
Root had faced a similar problem the year before in the Rand-McNally Building in which he had cantilevered one massive girder at the foundation to pick up and move the load of the party wall away from the existing footing of the neighboring Insurance Exchange. Fortunately for the budget of the building’s owner, the interior lot on the western edge of the site did not have a building, so Root could use a less expensive masonry bearing wall. If there had been a significant existing building on this lot, Root would have had to cantilever this foundation as well, and then could have claimed to have been “the first” to design an all-skeleton framed skyscraper. So my conclusion to this issue that has vexed historians for over a century is: Jenney in the Manhattan Building was the first Chicago architect to design a skyscraper using only the iron skeleton frame (but remember, New Yorker Bradford Gilbert had designed the first iron skeleton-framed building, the Tower Building, whose permit was approved only a month after the first publication of Buffington’s “Cloudscraper.”)
Bradford Gilbert, Tower Building, 1888. Note the five-story continuous piers in the arcade at the middle of the façade; Right: Diagram of diagonal bracing. (Landau/Condit, New York)
FURTHER READING:
Hoffmann, Donald. The Architecture of John Wellborn Root. Baltimore: Johns Hopkins University Press, 1973.