The builder of record for our house is Rowland & Rowland, father and son team Volney and Hermon Rowland. They began working in Berkeley in 1924, possibly looking for work after the September 1923 Berkeley Hills Fire.
The core of our house mystery (“Maybe a Maybeck”) is a couple of Rowland & Rowland houses that appear to be Maybeck designs: the 1926 Reid House at 24 Northampton in Berkeley and our 1927 Kingsley House in Montclair (“The Lost Maybecks”). These were written about in a Berkeley Architectural Heritage Association tour pamphlet, though the author suggests that our Kingsley House might actually be designed by Rowland - the implication I believe is that Rowland had used his experience working for Maybeck in the ‘20s to “inspire” our house design. We believe, however, that our house is a real Maybeck design done off the record - and we also have the “proof” letter from 1951 that describes a visit to the house by Volney Rowland where he mentions it was designed by Maybeck. However it’s clear from the letter that Rowland put a lot of his heart into the house and his own touches on the carpentry which may be what throws people off from accepting it as a Maybeck.
Regardless of the debate on whether our house is a Maybeck design, I wanted to track the Rowland’s career in the Bay Area. The obvious question is “are there other Rowland houses that look like Maybeck designs?” The only way to investigate that question is to survey all the Rowland permits. I had a partial list from BAHA for Rowland permits only in Berkeley. One of the remodels on that list looked like a Maybeck remodel on one of his older designs. It was, like our house, built by Rowland & Rowland with no architect listed. I started to think there were actually “2.5” lost Maybeck designs. Then I wondered about why this might happen that the true architect was left off the paperwork. I also looked at all the houses listed under Rowland permits to see if there was evidence that Rowland was “influenced” by Maybeck in his other designs - to see if he was starting to adapt some of Maybeck’s design quirks. Besides the “Lost Maybecks” which have a bunch of mid-20s Maybeck specific touches, there didn’t seem to be much influence on the other, simpler houses that do actually look like Rowland designs. So if Rowland was copying Maybeck, he did an excellent job in 1926-7 on these two “lost Maybecks” but didn’t copy Maybeck’s style on his own designs before or after ‘26-27. To me it looked more likely these houses are real Maybeck designs that are not officially attributed to him. Maybeck designs are very difficult to imitate because they are site specific and had a lot of unusual features that rotated and changed over his career. One would really have to be an architecture student of sorts to pick up on the design principles and stylistic touches that would make a proper mid-‘20s Maybeck floor plan and exterior appearance (a basic set of plans). Only half of the design is on the building plans, and the beautiful interior touches, personally done or supervised by master artisan Maybeck, would add further difficulty to any attempt at a Maybeck copy.
In my research I came across some biographical information about Rowland in a book about his Klondike days. Volney Rowland’s formal education ended with the 3rd grade. The Rowlands were not architects - they were builders... so if we take the permits at face value, it seems pretty surprising that a builder with a very minimal formal education would be able to make such spot-on-yet-completely-unique late-period Maybeck “copies”.
Judging a likely Maybeck design is less subjective than you might think - and in my opinion it doesn’t require a PHD in Architectural History - you really just need to read through the books about Maybeck. There are certain features that were Maybeck touches that would be highly unusual to see on average buildings from the ‘20s. One is a large setback on the lot - many Maybeck hillside designs have the front of the house about halfway back on the lot. Where most houses might be set back 20 or 30 feet for a front lawn, our house is about 75 feet from the street. The front door is still another 20 or so feet from the front of the house, with the door facing perpendicular to the street. This is an impractical choice for the older woman who commissioned our house, but does fit into Maybeck’s Hillside Design philosophy. The Reid House has a very similar large setback to a living room with a private entrance on the left side far back from the street. The Kingsley and Reid houses also both have a stone-faced garage on the left front corner of the lot - not seen in other Maybeck designs, this may be a Rowland touch.
I my research I came across the realization that the easiest way to tell a Maybeck design is his signature Venturi chimney. Nearly all Maybeck designs after 1907 have it, and the Rowland “designs” in question have it too. It’s my belief that only Maybeck designs feature this unique chimney.
Reviewing the list of permits from Building and Engineering News (BaEN), the first appearance of Rowland in the Bay Area is in 1924. In 1924 he is listed on 3 house permits on Le Roy Avenue in Berkeley. (It’s hard to know the exact order because the permit is only roughly correlated to the actual start of construction - plus or minus a couple months I’d guess). Sometimes completion notices were published in BaEN but usually not. For my purposes I will assume the houses were built in the order that they appear in BaEN.)
The “second” house he built on Le Roy was for Lilian Bridgman, a Maybeck protégée. My best guess for how Rowland met Maybeck was Maybeck lived nearby and walked over to check out the project.
However they met, after the three houses on Le Roy, Rowland and Maybeck were first working together officially on the Geisler House (still in 1924). Their first project is their ONLY official project together, even though it’s known they were in close partnership in the ‘20s.
In 1925, Rowland built another Maybeck & White, but this time no architect was listed. I view it as a transitional project between the formal first project and the completely informal projects that came after.
The “one room house” for Warren P. Staniford was one of several one room houses designed by Maybeck when he was experimenting with open floor plans. There’s an article about his ideas in Sunset Magazine from July 1923 (“The Maybeck One Room House”).
So how is it that a known Maybeck & White design has a permit with no architect listed? It’s not as straightforward as you would expect - not all of a given architect’s designs would be recorded at the building department because it’s a matter of bureaucracy and paperwork. The official list of Maybeck works was published by Kenneth Cardwell in 1977 in the first major retrospective book about Maybeck. Cardwell compiled this list from Maybeck’s office records and interviews with the architect himself. Cardwell makes it clear that the official list could contain errors but later authors seem to accept the official list as the best that we can know. So if we know that the official list could be incomplete, then how and why isn’t Maybeck listed on certain permits? My theory is that the name on the permit is usually the person who pulls the permit. Usually this would be the architect, but I believe this is in effect the “point person” for the job who has to deal with paperwork and inspections. If Maybeck trusted Rowland to take the project to completion, he could have Rowland pull the permit and it would reduce his workload of site visits and paperwork. At the bottom of permits at the time it says “signature of architect, owner or builder.” I think who ever goes to the building department and signs the permit application would be in charge of the project. It’s possible if an architect was listed, then they would be expected to be in charge of the project.
The next two projects I think were a collaboration between Rowland & Rowland and Maybeck are the 1926 Reid House and the 1927 Kingsley House - “The Lost Maybecks”. BAHA suspects the Reid House of being an informal Maybeck, and we have the letter for our house. If you believe my theory that the 1925 Staniford house was transitioning to a less formal collaboration, then these Lost Maybecks are completely informal - no office records that I have found mention these two projects at all. If there were records in the Maybeck papers, they would probably have mentioned these projects in one of the 3 main Maybeck books. But these two “Lost Maybecks” do look like Maybeck designs, and fit in with a pattern between Rowland and Maybeck of a working relationship where there was no mention of Maybeck on the permits. Some of these are known as Maybeck designs after the fact - but not from the permit record.
Another project of note is the 1926-27 Keeler House remodel. The Keeler House was one of Maybeck’s first residential commissions which stands just North of UC Berkeley. The remodel in 1926-27 subdivided the house, and its exterior seemed to be updated to Maybeck’s ‘20s style with stucco exterior and signature Venturi chimneys. I considered this another Lost Maybeck design but the Mark Wilson book about Maybeck has it as an official design so there must be a record of it somewhere in the Maybeck papers. It seems Mark Wilson had access to additional Maybeck papers through Jacomena Maybeck and maybe that’s where he found out about the Keeler remodel being a Maybeck design. Again it’s a Rowland & Rowland permit with no architect listed.
Another project worth mentioning is the addition to Maybeck’s 1924 Sack House. Maybeck’s own house burned down in the 1923 Berkeley Hills Fire. (It’s speculated by Mark Wilson that Maybeck started using almost exclusively fireproof siding and roofing after his 1915 Mathewson house (stucco exterior) survived the Berkeley Hills Fire while everything around it burned down). In 1924 Maybeck built a small house with a novel approach to siding - he used burlap sacks dipped into “bubblestone” (an aerated concrete product he was developing) and hung them by wires to create an effect like shingles. In 1926 the Sack House became the Maybeck Studio when an addition was built by Rowland. I don’t have access to this permit but I suspect there was no architect listed. Again my theory is that if Rowland was getting the permit, Maybeck would ask that his name be omitted from the permit card to avoid the paperwork. But I think it’s obvious that Maybeck designed the addition to his own home.
Another project of note, this time NOT by Rowland is Annie’s Cottage, 1933. It has distinctly Maybeckian features and is a known Maybeck but Bernard is listed nowhere on the permit - only Annie Maybeck. Again my theory is that Annie was going to the building department so Bernard was omitted from the paperwork. He just did the drawings. I think Maybeck enjoyed drawing and would have been content only doing the drawings for some projects.
I believe the last house of this Rowland and Maybeck collaboration was our 1927 Kingsley House. I can only speculate on the reason for this. Rowland was around retirement age in 1929 when the stock market crashed and he moved to the Central Valley (to “retire” and breed silver foxes). It does seem like work was drying up before then as the number of permits under Rowland was steadily declining from 1924-29. Maybeck better weathered the economic downturn with a couple large, ongoing projects - the Earl C Anthony mansion in LA and the Principia College plan in Illinois. Our house took exceptionally long at 14 months (from our letter) and went well into mid 1928. (Most houses during this time I would guess took around 6 months). After our house, Rowland next in 1928 worked on another house in Montclair. Rowland was doing about 3 houses a year from 1924-26 but only 1 or 1.5 from 1927-29.
Rowland finished his last house mid 1929 at 36 Northampton (next door to the Reid House.) it’s possible that the two friends simply didn’t have time to work together anymore after our house. I believe Maybeck just made the basic building plans, maybe 6-10 pages, for these unofficial collaborations and the rest was up to Rowland. Maybeck earlier in his career was very involved in all the fine details of the construction - sometimes actually doing some of the woodwork. After he was hospitalized briefly from a health scare potentially from overwork, Annie reorganized his office in 1924 and hired draftsmen to reduce his workload so he could focus on what he enjoyed - drawing “pretty little pictures”. Scholars note the decline in the quality of his work after this point, when he could no longer obsess over every square inch. With Maybeck not available, Rowland presumably had to fill in some details like the carpentry and tile work. Maybeck building plans like others from the time were rather sparse. Everything not on the plans would have to be filled in by the client and contractor. The Rowland touches combined with Maybeck’s plans make the Kingsley and Reid houses two of a kind, unlike any Maybeck before or since.
Of the 15 projects Rowland completed from 1924-29 in the BaEN, only 7 don’t have an architect listed meaning they were listed as “Architect-None” or the architect was listed Rowland & Rowland or V. H. Rowland (Volney Rowland). Of these projects, many are known or suspected Maybeck designs. There are a couple that look like actual Rowland designs. These are relatively simple and would not be difficult for a contractor to design. One is 2031 Del Norte, which is a simple L shaped house on a corner type lot with a front door in the middle of the L. The simplicity of the form combined with the front facing door make it clearly NOT a Maybeck. The contrast between these contractor designs and the suspected Maybeck designs show that Rowland was a humble contractor, not a master at architectural forgery. At this point in my research, it seems much more plausible that Maybeck designed some projects for Rowland off the record, than Rowland somehow understanding Maybecks hillside design philosophy so completely that he could implement his ideals to unique lots with complex interactions between the building and the site it sits upon.
Below you can see my “Rowland” search of Building and Engineering News turned into a handy spreadsheet. The top is all the permits listed under the name Rowland. If you take away all the projects with no architect listed or Rowland as the architect, you have 5 houses and two remodels ostensibly designed by Rowland. (You need some type of design to build a house, and no architect is almost the same as contractor-designed though I supposed the client would have some input. Even with a client sketch, the builder would need to decide on the actual measurements and construction.)
Within the “Arch:Rowland/None” group falls the 2 “Lost Maybecks” (yellow) and 2 known Maybeck designs that weren’t credited on the permit (green). Before and after what appears to be an off the record collaboration, are 2 houses and a remodel that look like contractor designs - like Rowland actually did design these (blue).
In summary, I think the actual Roland/Maybeck collaboration is as follows:
1924 Geisler House, Maybeck and White
1925 Warren P. Staniford House, (Maybeck and White)
1926 Keeler Remodel (Maybeck)
1926 Reid House (Maybeck?)
1926 Maybeck Studio (Maybeck)
1927 Kingsley House (Maybeck?)
If you take this one step further, you can try to calculate what percentage of Maybeck’s likely designs were off the record. If we have 4.5 house designs from the Rowland & Maybeck collaboration, then add Annie’s cottage, the Aikin house, another house I was told about on a tour, a concrete building in Berkeley - that’s about 8 or maybe more out of a total of 180 designs - that’s only about 4% of his total works. Since Maybeck definitely has off the record designs - it seems possible that there may be more lost Maybeck designs. This article is a “proof of concept” that the subject warrants further investigation.