Buildings convey meanings. 1 People make inferences from building exteriors about what goes on inside 2 3 Should a courthouse look more like a courthouse than a factory, prison, or office building? If so, what form would that take? 

Until the rise of modern architecture in the 1930s, public buildings and courthouses in the U. S. often had a neo-classical style. 4 Buildings in this style, such as the White House and U. S. Supreme Court building, tended to be white, and they had symmetry, columns, and a triangular pediment. People recognized them as public buildings. With the rise of modernism, architects abandoned the neo-classical forms. Not only does the public dislike the appearance of such modern designs, 5 but such designs disconnect form from function; from the outside, people cannot tell what the building houses. 6 Critics have described these modern courthouse designs as “over-scaled” complexes that “intimidate people” 7 and noted that the designs raise questions about what makes a building look public. 8 

For courthouses, the U. S. General Services Administration (GSA), which oversees the design and construction of federal courthouses, called for them to appear open—that is, responsive, transparent, and accessible to the public. 9 Although GSA has overseen billions of dollars of federal courthouse construction in most states, it has neither studied nor codified how to make the courthouses look open, appealing, or like a courthouse. In short, while spending billions of our tax dollars, it does not know what a federal courthouse should look like.  

GSA does have a Design Excellence Program that hires distinguished architecture firms and panels of distinguished private architects to review proposed designs, but such panels are consistently wrong in judging likely public impressions of buildings. 10 Consider one of many studies that considered this question. An evaluation of three federal courthouses found that the design features used by architects to convey openness had nothing to do with the features that occupants saw as conveying openness. 11 To find out what courthouses should look like and what features they should have, I conducted a small study using methods standard to empirical studies of environmental aesthetics. The study confirmed that old (neo-classical styled) courthouses look more like courthouses than do the newer (modern) ones.  

I had a diverse sample of 28 people respond to color photos of 36 buildings (18 courthouses and 18 other public buildings such as libraries and museums). The buildings varied in age (built before 1900 or after 1980) and size (one-to-two stories, three-to-four stories, more than four stories). Interviewers shuffled the photos and presented them to each participant. Participants were asked to sort the buildings into five piles, once for the degree to which each one looks like a courthouse, and once for the degree to which it looks open to the public. As these two responses were similar, I combined them into one “looks-like-a-courthouse” score.

Age and size mattered. The older buildings looked more like a courthouse than did the new ones. The large- or medium-sized buildings looked more like a courthouse than did the smaller ones.  

To understand what specific characteristics make a building look like a courthouse, I arranged the photos from the highest to lowest score and inspected their physical characteristics. This revealed that the presence of four neo-classical characteristics made a building look more like a courthouse. The higher scoring buildings had: 

  • Columns
  • Pediments (extended forward)
  • White color, and
  • Symmetry.

The findings agree with consistent findings that architects misjudge public likely public impressions of a design, and that most non-architects dislike “modern” design and have done so for almost a century. Perhaps, through repeated experience, U. S. citizens have learned to see neo-classical designs or designs with those features as good examples of public buildings such as courthouses. This does not mean a courthouse must have a neo-classical design. Instead, it suggests that a design should incorporate the four neo-classical properties of columns, pediments (extended forward), white color, and symmetry.

How accurate are responses to color photos by this sample? Studies indicate that they are very accurate. An analysis of responses by more than 2, 400 people to more than 150 environments found that responses to color photos accurately predicted responses on site. 12 An analysis of responses by more than 19, 000 people to more than 3, 250 environments found high correlations between groups that differed by age, gender, education, culture, and interests, and it found that the effects of the physical environment were magnitudes larger those of individual difference among respondents. 13 Of course, responses on site to buildings by larger samples responding could show how well the predictions hold. GSA might also do well to consider studies of responses to other kinds of public buildings, such as embassies and consulates, to find out the meanings they convey. Such work could help shape designs to better convey the intended meanings to the public. If the present findings hold, they suggest that the best designs will incorporate some traditional forms and features.


Prof Jack L. Nasar is the Academy Professor of City & Regional Planning, The Ohio State University, and former editor of the Journal of Planning Literature

  • 1. Rapoport, A. (1990).  The meaning of the built environment: A non-verbal communication approach.  Tucson, AZ: University of Arizona.
  • 2. Sadalla, E. K., Vershure, B., & Burroughs, J. (1987). Identity symbolism in housing. Environment and Behavior, 19(5), 569-587.
  • 3. Farbstein, J., & Kantrowitz, M. (1986). The image of post office buildings: First findings—focus group.  The Costs of Not Knowing--: Edra 17/1986: Proceedings of the Seventeenth Annual Conference of the Environmental Design Research Association,  Atlanta, GA., 17259.
  • 4. Walton, T. (1992). Pride and stewardship: Renewing the mandate for design excellence in America’s public realm. Places, 9, 4-19.
  • 5. Devlin, K., & Nasar, J. L. (1989). The beauty and the beast: Some preliminary comparisons of ‘high’ versus ‘popular’ residential architecture and public versus architect judgments of same.  Journal of environmental psychology,  9(4), 333-344.
  • 6. Nasar, J. L., Stamps III, A. E., & Hanyu, K. (2005). Form and function in public buildings.  Journal of environmental psychology,  25(2), 159-165.
  • 7. Phillips, T. S. (1999). Courthouse buildings: Designing justice for all. Architectural Record, 187, 105-107.
  • 8. GSD News. (1995). Architecture in the public realm: A public discussion. GSD News/ Harvard University, Graduate School of Design,  (Winter), 20-27.
  • 9. GSA. (1999).  Balancing openness and security: A thematic summary of a symposium on security and the design of public buildings. Washington, D. C. : Public Services Administration.
  • 10. Nasar, J. L. (1999).  Design by competition: Making design competition work. Cambridge University Press.
  • 11. Pati, D., Rashid, M., & Zimring, C. (2010). Occupants’ perceptions of openness in federal courthouses.  Journal of Architectural and Planning Research, 27(3), 257-269.
  • 12. Stamps III, A. E. (1990). Use of photographs to simulate environments: A meta-analysis.  Perceptual and Motor Skills,  71(3), 907-913.
  • 13. Stamps III, A. E. (1999). Demographic effects in environmental aesthetics: A meta-analysis.  Journal of Planning Literature,  14(2), 155-175.